bright
bright
bright
bleft
bright
bleft
bright
tl
 
tr
lunder

Reviews of Language Testing Resources on the Internet

This site designed and maintained by
Dr Glenn Fulcher

@languagetesting.info

runder
lcunder
rcunder
lnavl
Resources
lnavr
navtop
lnavs
rnavs
lnavs
rnavs
lnavs
rnavs
lnavs
rnavs
lnavs
rnavs
bottomnav
lnavl
Information
lnavr
navtop
lnavs
rnavs
lnavs
rnavs
lnavs
rnavs
lnavs
rnavs
bottomnav
lnavl
Tools
lnavr
navtop
lnavs
rnavs
lnavs
rnavs
lnavs
rnavs
bottomnav
 
 

 

 
Index
     
   
     
 
Reviews
     
 

Introduction

The Language Testing web site is designed as what Geoff Brindley has termed "a one stop shop" on the Internet for language testing practitioners and students. It started life in early 1995. The links and associated reviews were the orignal part of the site. All the reviews on this page relate to the LINKS that you can access in the left hand navigation pane. Where a link has a review, by clicking on 'review' you will move the relevant part of this page.

The list of resources is the result searching the Internet and including those resource that are brought to my attention by users of the site. Extending this list of resources is an on-going process, and this review document is changed with every update of the page. However, while I add new material, I do not delete older material. In this way it has become something of a graveyard of lost resources, as well as a review of what is currently available. This makes it a very strange document. So I have to extend my apologies if in places it reads like the constantly edited monster it has grown into.



Click here to go to the top of the document

Language Testing Associations (with individual membership)

On 1 May 1996, I established a link to the International Language Testing Association (ILTA) Homepage. Initially launched with information on its purpose, its officers and committee members, membership, and a request form. To quote from the ILTA pages: "The purpose of this organization shall be to promote the improvement of language testing throughout the world." Anyone who takes a professional interest in language testing should consider membership. ILTA is the umbrella organization for the profession. It has led the way in introducing an internationally agreed Code of Ethics, and Practice Guidelines. Professor Alan Davies has done much of this work, which has been much appreciated by the entire field. It is sometimes the case that ILTA works slowly, and it has been criticised for this. But it is inevitable if language testers across the world are going to agree to common values and practices within what is essentially a very democratic organisation. ILTA also organisers the annual Language Testing Research Colloquium, and finances workshops on language testing around the world. It has grants to enable students to travel to LTRC, and a number of other awards that recognize achievement in the field.

I managed the ILTA web page from the Universities of Surrey and then Dundee, until an independent web site was established in April 2005, and my resources page was decoupled from ILTA. In 2008 a new web site design was unveiled, designed by a professional designer from Teachers College, and managed by Prime Management. This is an excellent site with a great deal going for it. There is a membership only area, which contains some ILTA publications, including the constantly updated bibliography of language testing.

The East Coast Organization of Language Testers is another professional organization or language testers in North America. They organize an annual conference and other activities. Their site gives details of the next conference.

The Japan Language Testing Association added in July 2000 (although I'm sure this has been around for a lot longer than that) is mostly in Japanese. However, the English sections explain the aim of the association: "to improve both the th eoretical understanding and practice of measurement and evaluation in foreign language education in Japan" - and the relationship between ILTA and JALTA certainly led to a very successful LTRC in 1998. They continue their work through dissemination of research in Japan, and organising conferences and workshops. If you live in Japan and are interested in testing, you really ought to join!

The Midwest Association of Language Testers is an inclusive organization for professionals based in the American Midwest who are interested in Language Testing. The organization says:

MwALT shall serve to:

  • Stimulate professional growth through workshops and conferences
  • Promote the dissemination of information related to language assessment
  • Foster opportunities for training in language assessment
  • Provide leadership in language assessment
  • Offer professional services to its constituencies Increase public understanding and support of language assessment
  • Recognize outstanding achievement in language assessment
  • Establish links with other groups interested in language assessment

Launched in 2003, the European Association for Language Testing and Assessment is seen as a forum for the discussion of language testing issues in Europe. The rationale for this organization is that Europe has specific assessment issues to deal with that are unique. In particular:

"....developments in Europe, particularly associated with language policy in the European Union and the Council of Europe, including the dissemination of the Common European Framework and the European Language Portfolio, have highlighted the need for a European language testing association, which speaks for testers in Europe and which has no commercial affiliation or aims.

When I first established the link to EALTA, I wrote: "I hope that this also includes 'no political affiliation or aims', but this remains to be seen. Unfortunately, in Europe, much of what appears to be funded and approved of is that which the great central system thinks will lead to even greater unity and hence more centralization. In ideology, research doesn't really play a significant role unless it supports the conclusions you've already reached." And EALTA was founded with EU financial support.

Over time EALTA has shown itself to be very different from other regional organizations. In other associations the regional name merely indicates where the membership is drawn from; in the case of EALTA the regional name implies identity, and support for the uniqueness of the regional approach, complete with its own texts, to which there is clearly an expectation of some adherence. In 2009, without consulting or seeking a mandate from the membership, it became 'recognized' as a non-governmental organization by the Council of Europe, and EALTA publications began to include statements like "EALTA endorses the values, aims and objectives of the Council of Europe". These may or may not be laudable, but the organization did not have a mandate to endorse these political objectives on behalf of its members. It also means that EALTA is not an independent professional organization. In short, my hope that it would have 'no political affiliation or aims' has been dashed.

The organization has done some good work, as identified on the web site, including its annual conference and the creation of guidelines for good practice. However, its rather narrow European interests make its relevance to language testing more generally somewhat questionable. It has even become parochial, rather than regional, as a result of its dogged devotion to European institutions. I was an 'expert member' for a number of years. Finally the reluctance of EALTA to allow all members to have a democratic say in major policy decisions meant that I could no longer in all good conscience remain a member. I resigned from the organization on 17th February, 2010 after being told on the EALTA discussion list that "democracy is a weasel word". These are 'values' which I certainly do not share.

Test Providers

A newcomer in the 1 April 1996 update to the list of sites relating to language testing was the Association of Language Testers in Europe (ALTE) at Edunet. It disappeared in early 1998 without trace, and reappeared in 1999. I added a new link on 9th July. ALTE is an organisation consisting of various language testing bodies in Europe, their purpose being to

  • establish common levels of proficiency in order to promote the transnational recognition of certification in Europe
  • establish common standards for all stages of the language testing process....
  • collaborate on joint projects and in the exchange of ideas and know-how.

The ALTE web pages describe the work of the organisation, and provide a useful glossary of the terms used on the pages. They provide a copy of their Code of Practice, and definitions of the various levels they use to describe the tests. A nicely presented web site.

Other ALTE members to whom we now have links are the Deutscher Volkshochschul-Verband, Institutionen for nordiska sprak at the University of Stockholm, Alliance Francais and Instituto Cervantes. Each site tells you what kind of tests the institutions offer, and when and where you can take them. At my last visit to the Deutscher Volkshochschul-Verband just before 3rd July 2000, I noticed that they had expanded their site to include some practice tests, but apart from this, these sites are pretty dry.

Quite amazingly, I only came across the Association of Test Publishers by chance, and added the link to their website in March 2010. However, this organization has been in existence since 1992, and as the website says, it "is a non-profit organization representing providers of tests and assessment tools and/or services related to education, employment, certification/licensing or clinical uses." The membership is only open to organizations, clearly. They run an annual conference, and various other events for their members, some of which look very interesting, so it's a pity they aren't open to individuals. However, in this case the focus is clearly upon the interests of the test providers, and so the kind of critical approach you would find at LTRC, for example, may very well be absent. But this is merely me surmising - I really don't know. They have their own journal (of applied testing technology) that I'd never heard of. I tried to find this through my university journal search, but it wasn't available. So I've no idea what it's like. But they do produce their own newsletter, which is freely available for download. You can judge for yourself. There is also an interesting FAQs page with information on testing - most of which has a positive "pro-industry" slant - which is of course not surprising. Most of the web site is available only to members, so I can't comment on what's there.

Cambridge ESOL have a very extensive site that has grown over the years. It covers all of the main suite examinations, plus many other tests that they offer. For learners they have pages explaining which of their tests might be the most appropriate. The navigation panes take you effortlessly to full descriptions of the tests, handbooks, and sample papers. For teachers there are also links to reports on the tests, and links to various materials that will help in preparing students. For researchers the section explaining research at Cambridge and the successful Research Notes series is extremely useful. However, it is unfortunate that the site does not have the full research reports that are provided on the sites of some other test providers. That said, this is an extremely well-designed and informative site.

Educational Testing Service maintained a gopher site until late 2000, but now only offers a World Wide Web site. Their original WWW site was appalling, and it thankfully disappeared in December 1995. The site which re-emerged in July 1996 was excellent, and the 2002 redesign even better. Since then it has morphed through a variety of designs. It is now extremely easy to navigate and very user friendly. The only thing that really irritates a user like me is the length of the URLs, and the frequency with which they are changed. It makes linking to information extremely difficult. Apart from providing extensive information on ETS tests, including samples, ETS provides an extensive range of research documentation that is extremely valuable for teachers and researchers in this area. Much of that relating to language testing has been conducted in relation to TOEFL, and you should follow the link to that test in the section on 'specific tests'. ETS also provides a test finding service, and you can find a link to that under the 'search for tests' section. All in all, this is a great site with a lot of information that is useful and relevant for learners, teachers, and researchers.

The The ISLPR was originally based in the Language Testing and Curriculum Centre from Griffith University, is where you could, from time to time, find the International Second Language Proficiency Ratings, which grew out of the Australian Second Language Proficiency Ratings (ASLPR). The university changed the URL of its site sometime in early May 1996, and I was informed of the new URL in April 1997. Links re-established. But then they changed them all again sometime in 2001. I finally managed to find the pages again in September 2001. Then they disappeared again, and I searched and found them again. WHY DO THEY DO THIS? Finally, the link disappeared once more in mid-2008. However, with the help of a colleague in Australia, we traced the new URL of the ISLPR, which has moved to a new web site, and a new address, which inauspiciously is: Upstairs (door next to Dentist), Office Suites, Shopping Centre, Cnr Padstow & Warrigal Roads, Eight Mile Plains. You used to be able to download the ISLPR scales and all sorts of stuff. Now you can buy them, and book a test.

Pearson Language Tests is part of the Pearson publishing conglomerate. Currently they offer the London tests of English, and the London tests of English for Children. Under preparation for launch in 2009 is the Pearson Test of English, which is designed to be a test of English for Academic Purposes that will attempt to challenge IELTS and TOEFL as the two key tests in this particular market.

Second Language Testing Inc. is a company set up by Charles Stansfield to provide test development, translation/adaptation, and other related testing services. SLT develops tests for a wide range of clients, and is one of the few providers to specialise in aptitude testing. The section of the web site that explains aptitude testing is excellent, and provides useful references. This is a great site to explore.

Click here to go to the top of the document

Specific Tests

BULATS is the Cambridge ESOL computer-based test of business English. This web site tells you this, and what you can use the test for. Potential candidates can download a handbook, and there's also a test demo. That's it in a nutshell.

Information on the International English Language Testing Service (IELTS) comes from Cambridge ESOL. This site has improved immensely since I first reviewed it. In August 2007 it had some interesting flash built into it, but the content was what really made the site different. You can now download a test handbook and samples. But more important is the teachers' resource section, which provides not only an overview of what is on the test, but ideas (and example materials) for how learners can prepare for the test. Also available is data on learner performance on the test, and some basic statistical information on test performance. This is only provided for the objectively scored parts of the test, but despite this limitation it is nevertheless to be welcomed. The advertising is a little over the top, but it's discrete enough not to be distracting from the useful content.

Developed by Eurocontrol (why on earth must every organization in Europe now begin with 'Euro'?) to certify pilots' and air traffic controllers' abilities to communicate in English to a level that ensures safety in the air. This is clearly very high stakes testing, and pilots or controllers who do not pass the test can have their licenses withdrawn. This is clearly one area where the very highest standards of test design, development, validation and administration, are paramount. There should never be any corner cutting with high stakes tests of any kind, but it is with the certification of communication skills where errors can be fatal that its importance is brought home to bureaucracies. Needless to say, because it is high stakes, it is also an area of controversy – so a site that is well worth visiting to see what is being done in Europe.

The OET was originally developed by Tim McNamara at the University of Melbourne and is one of the best researched tests of language for specific purposes. This is a site well worth visiting, and anyone interested in the design of the OET should read Tim's excellent book, published in 1996 with Longman (now Pearson Education) entitled Measuring Second Language Performance.

The link to the STEP Test was added on 6th January 1999. In March 2006 I noticed that there was an English version of the site, which was particularly welcome, as the STEP test is an important test in Japan and has attracted much attention from around the world. The site just described the test and its levels, but not in any great detail. In March 2010 I received notification from STEP that they'd completed an update of the site, and intended to expand it further in the future.

The new web site is certainly attractive. The designers have gone for a simple design, with lots of white space. Visitors are not going to get lost here, and the interface is easy to navigate (just beware that your cursor doesn't always turn into a hand when you hover on a clickable link!). There's news on the home page, which I assume is driven by an RSS feed; but you can't tell from the source code, of course. There is now much more information about the test, including downloads of sample papers and audio (instructions/listening). These are very useful for potential test takers, as well as anyone interested in finding out what the test is like. I'll certainly use these with my own students when I teach them how to evaluate tests. The potted history of the test is also helpful to understand its evolution.

There is also some information on research outcomes. On my last visit (11th March, 2010) I read the outcomes of a can-do study, which attempts to match student reports of what they can do with the language (perceptions, of course) to score grades. There is also a 'research' section, where language testers can find references to the relevant research papers that discuss these projects to support score meaning and interpretation. No links though, even where these are available online. Although couched in the language of academic validity theory, it is not surprising that all the efforts are aimed at building a validity argument to support the intended use of the tests; this is quite normal for testing agencies that wish to increase test volume. The game at the moment for most testing agencies is trying to demonstrate linkage to external (powerful) frameworks or tests. In this case, it is the CEFR and TOEFL. The purpose is clearly wider recognition, not only among academic institutions, but the widest possible range of employers. Can one 'proficiency test' fulfill all these needs? Clearly not. But there is a slide among all testing agencies to treat 'validity' as 'recognition', and the CEFR encourages this subversion. (I am arguing this in a symposium at LTRC 2010 in Cambridge organized by Bernard Spolsky, entitled "The Industralization of Testing". This is linked to the continuous expansion of test purpose, and our need to 'contain' testing agencies to their original purpose unless a retrofit program and argument is followed through. See: Fulcher, G. and Davidson, F. (2009). "Test Architecture. Test Retrofit." Language Testing 26, 1, 123 - 144; Fulcher, G. (forthcoming). "Test design and retrofit." In Chapelle, C. (Ed.) The Encyclopedia of Applied Linguistics. Oxford: Blackwell.)

It's clear that a number of external researchers are involved with the test, and this will help the STEP to develop research that is not entirely market driven. I will look out next to see if STEP institutes a program of research that funds independent researchers to investigate validity questions with EIKEN data, perhaps with a focus on some of the alternative hypotheses that make up the 'rebuttals' to the primary validity claims. After all, anyone who looks for a relationship of their test to the CEFR is going to find one. It hardly constitutes validity evidence of any value at all, however much it might help with recognition in Europe and test volume.

My view? The new site is great. I'm sure the quality and range of information will grow. If it's got me thinking about what they're doing, where they might go, and interested in the research conducting (however critically - that's what I'm here for), things are moving in the right direction.

The TOEFL offers a high quality site with the usual test dates and registration information, online practice tests, and information about the nature of the test for teachers and learners. But it goes a long way beyond this. In the download library there is extensive information on performance on the test, as well as test statistics. What is most useful is that TOEFL provide the scoring criteria for the iBT along with sample responses. If an institution wishes to set it's own TOEFL iBT levels for entrance you can also ask for a standard setting pack that contains a detailed description of how to carry out your own study, along with additional benchmark samples. For language testers, there is also a very interesting section on research activities. This has an overview of TOEFL research, information on new research, an account of the TOEFL 2000 project, and a list of TOEFL Research Report and Technical Report series. The TOEFL Research Report and Monograph Series has become a very source of information, not only on research into TOEFL, but on the development of innovative research methodologies. This site comes highly recommended.

The official TOEIC Page is from ETS. There are many other TOEIC pages out there though! So be warned. From around 1996 to 2003 TOIEC was part of the Chauncey Group International (a limited company, wholy owned by Educational Testing Services), but it has now been incorporated back into ETS proper. What is TOEIC for? Well, they say:

  • For language training programs and schools
    • Placement
    • Measure of student learning
  • For companies and other organizations
    • Selection
    • Promotion
    • Measure of employee readiness for English-language training
    • Vehicle to motivate staff learning
    • Tool for benchmarking

The rest of the site amounts to a simple description of the test, and the usual contact details for test takers. When I last visited in August 2007 the research reports had disappeared, which is unfortunate, but they had added a handbook for test takers, a can-do guide, and a number of other documents.

Click here to go to the top of the document

Personal Pages

This new category was added on 1 May 1999 to accommodate a single entry: The Foreign Languages in Tourism Training page which was added on 1 May 1999, and it had disappeared by August 2006. I don't know who authored this page, so I can't find out where it has gone. It was a wonderful resource while it lasted. This site provided a lot of information on the ALTE framework, and also provided the full ACTFL Guildelines for the reader to compare against ALTE band level descriptors. There was also information on the LCCI English for Tourism Industry examination. A great pity that this site ceased to exist.

Maintained by Dr Helen Barrett, this web page contains links to a variety of information on so-called 'alternative' and portfolio assessment. This has to be one of the most comprehensive set of links on this very specific area of testing and assessment.

Maintained by Franz Rubenbauer, this is a fascinating web site dedicated to aviation English and the problems that arise when people working in the industry fail to communicate. I first came across this in July 2009. The purpose of the web site "...is to provide information on erroneous English communication within aviation from a linguistic point of view and with the background of pending worldwide language testing." And it does a terrific job. There is a list of communication failure related incidents, a summary of the author's own research in the area, as well as various glossaries and links. I was particularly interested in watching the embedded YouTube movies. It makes you wonder about the safety of flying! Great site. worth spending time here to realize just how important language testing can be in high stakes contexts.

Click here to go to the top of the document

Government Sites

This new category appeared on 1 November, 1997. Although I am aware that a number of the organisations listed in the private category are funded by the US government, they are still NGOs.

The Defense Language Institute English Language Center provides an English test for military students. There is a description of the test, and you can also take sample items. The site says:

    The ECL is the Department of Defense test for assessing the English language proficiency of international military students being considered for assignment or training in the United States. It is used in more than 110 countries, as well as the United States. The required ECL score for course entry is set by the military service and school which conducts the course, not by DLIELC. The test is scored on a scale of 100, and most requirements fall between 65 and 85.

The National Tests is provided by the US government. I am reliably informed that this is a hot topic in the US. From the page, I quote:

  • In his State of the Union Address on February 4,1997, the President named education as his "number one priority for the next four years." He also issued a call to action -- a 10-point plan -- for ensuring that Americans have the best education in the world. The first point is:
    Set rigorous national standards with voluntary national tests in 4th-grade reading and in 8th-grade mathematics to make sure our children master the basics.

I'd love to comment, but as a Brit' I think it would be safer just to stay quiet on this one.

This is the web site of the National Assessment Governing Board. As they say, "The Governing Board is an independent, bipartisan group whose members include governors, state legislators, local and state school officials, educators, business representatives, and members of the general public. Congress created the 26-member Governing Board in 1988 to set policy for the National Assessment of Educational Progress (NAEP) -- commonly known as the "The Nation's Report Card." In November 1997, Congress granted NAGB exclusive authority over development of the proposed Voluntary National Tests (VNT)."

On 4 February 1998, I added another government site: the Consumers' Guide to Performance Assessment. This was recommended by a contributor to a debate on portfolio assessment on ltest-l. It wasn't particularly long, but provided a useful introduction. There were lots of contact addresses, too. But at the turn of the century, I guess the government decided it was no use, and took it away.

The Interagency Language Roundtable is now available on-line, and I added this link on 1 May 1999. This page not only provides the entire scale, it also has a brief history of how it developed from the early FSI rating scales, and there is a brief bibliography relating to this history. For those who are not familiar with the way in which the ILR scale grew out of the FSI, and was then exported into schools and universities in the United States, this is a good introduction.

Many thanks for one of my own students who brought this wonderful site to my notice in February 2000, and added to the links in the 31 March 2000 update. The Examinations Appeals Board, which I had never heard of before. Basically, if you think that your grade is unfair in a UK examination, you can appeal to this body only after you have exhausted the appeals procedures of the examination board that set the examination. And here's the crunch:

The procedures open to investigation include the full range of processes involved in and leading to the award of grades: the setting of papers; marking and moderation; the grade award; enquiries about results and appeals and administrative arrangements. The appeals process is not concerned with making judgements about candidates’ work in the examination since that is properly the responsibility of the senior examiner of the awarding body concerned.

In other words, it will simply ensure that appropriate procedures have been followed by the examination board, and perhaps "recommend" that these be changed, or a case go through the appeals procedure again. How this can ensure fairness I am not sure. But the UK examination boards and their qualitity control procedures do work in mysterious ways. And no one enquires too closely about how they operate.

Yes, it's Ofqual or the Office of the Qualifications and Examinations Regulator. The UK government sets up quangos to regulate almost everything these days. Ofcom is for communications, Ofsted for standards in education, Ofgem for gas and electricity, and so on. One of my favourites is Ofwat for water. It doesn't matter what they do, we have flooding in the winter and hose pipe bans in the summer. Bawdy linguistic jokes playing on the 'Of' are not uncommon. So it's only fair to ask what Ofqual does.

Well, it appears from what was on the site when I first visited in early 2009, that the role of Ofqual is to work with test and examination providers to ensure 'the best deal for the learner'. What does this mean? Well, in the introductory video we have a smattering of ideas. The first is that the public, businesses, colleges and universities, should have 'confidence' in the tests. Hmmmm. Okay. How would this be achieved? The speakers seem to have an intuitive understanding of what might create this confidence. The first one says:

"We're looking for a regulator who will ensure that the qualifications that my students take here are assessed well, accurately and fairly reflect their achievements."

The repetition of "accurately" throughout the video is particularly troubling. We know from the work of people like Ennis (check out the link to his article in the articles section of this web site) that the use of words like this confound and confuse issues of reliability and validity, even if the former is now seen as a facet of the latter. Even the person who appears to be the head of Ofqual makes a similar error:

"I want Ofqual to be an organisation that works in partnership with awarding bodies but always with an eye to why it's doing that. And the reason why we've come into existence is to ensure that the learner gets a good deal from the qualifications and tests which they take and that those qualifications and tests are as reliable and accurate as they possibly can be and that they feed back into the learner a good sense of achievement."

Having said this, it is clear that there is a general awareness that examinations should be 'for the learner', although what this means never really materializes beyond a rather vague appeal to arequirement for 'fairness'. One has to wonder if the Ofqual officials - or indeed the heads of examination boards - have ever read the fairness literature, studied the history and development of standards documents, or got to grips with validity theory. In fact, I started working through many of the pages on this site and the downloads available, searching for information on validity. It does crop up from time to time, but it is essentially introduced as a term with a single sentence gloss. And I quote: "The fitness of purpose of an assessment tool or scheme." This does not do justice to the richness of the validity literature.

Given that this is the case, we can still evaluate the site on its own grounds, rather than what it doesn't contain. It appears that Ofqual's main weapon in maintaining public confidence is the production of codes of practice for examination baords and monitoring how their practices meet these codes. You can download them from the site and evaluate them for yourselves. In fact, the documentation provided is quite extensive, and could form the basis for a consideration of just how testing professionals might have to change and adapt it to stand up to serious academic critique. You can judge for yourself, but this appears to me to be a primarily bureaucratic activity.

Ofqual also conducts research and, as you would expect, it is directed toward answering the regular criticisms leveled at tests in the media. Of particular importance in the UK is what is called 'test standards' - by which journalists really mean the consistency of score meaning and item/task difficulty over time. This is at odds with the use of the term elsewhere, of course. But we'll let that pass. Every year the press report that the examination boards are making tests easier so that the government can meet its educational targets, set out in terms of the percentages of learners passing tests at certain ages. And every year education ministers get up to say it's not true, and the press are just running down the very real achievements of learners. It's all so predictable, it's amusing. Similarly, the reports on this web site tend to conclude that all is well in the world, like this example from the executive summary of McGraw, B., Gipps, C. and Godber, R. (2004). Examination standards: Report of the independent committee to the QCA, piii: 'The committee’s overall conclusion is that no examination system at the school level is better managed.' Also notice the multi-semantics hiding in the word 'standards', and there's no attempt to disambiguate its various meanings in the text. So there we have it. Public confidence is assured. In all fairness, there is evidence of passion about broad (but little understood) notions of 'fairness' and 'putting the learner first'. But from my own reading of the material on this site in February 2009, there's little evidence of any real grasp of testing theory, or an understanding of the philosophy of educational assessment.

Since I wrote this review, the so-called 'SATS scandal' has hit the UK. Do check out this news story from the BBC. This is just one more example of how the government wishes to maintain complete control, and sees the various quangos as devices to cover up mistakes. And as I was enjoying my morning tea on 7th May, I heard an intriguing story on the Today Show (Radio 4) with the head of Ofqual being interviewed. The issue was: what happens if swine flu breaks out and it disrupts examinations through school closures. The response from Ofqual was that they have "tried and tested methods" (oblivious to the pun) to ensure that those who would have taken the examinations get grades that are "fair and accurate" by estimating them from other evidence. I for one would love to know how they do this, and what the predictive evidential basis is for whatever the practices are. But from that interview, and the information on the web site, I have to conclude that it's most likely done with mirrors.

There is an excellent site in Israel where information is available on the English Bagrut Examination. It used to be hosted by the Hebrew University, but has now been transferred to the Ministry ot Education. There is a full explanation of the format of the tests, complete with information for teachers on how to prepare students, forms for classroom based student assessment, and the criteria for grading the papers. It is important to realise that this site is for English teachers and not language testers, but I feel that it is an excellent example of how the Web is being used to disseminate appropriate information on a national testing system to the people who need it.

Hungarian Exams Reform Teacher Support Project.This is simply a great site to visit. The web site documents a project set up to reform language examinations in Hungary, and unusually, the web site makes available all the documentation from the project for anyone to inspect. This includes how the tests were developed, the guidelines given to item writers, rating scales, interlocutor frames, and so on. The project also produced textbooks to support teachers who are asked to prepare learners for the tests, and since these have gone out of print, they have been made available electronically on the site. In short, there's a lot here for people to rummage through, and the materials provide an excellent example of how examination reform can work with teachers and learners, rather than against them.

The Manual for relating language examinations to the Common European Framework is difficult to review at present, because it only exists in draft form. There is also a problem with reviews about the CEF at the present, as its status as a 'meaningful framework' is under debate. However, what I can say at the moment is that the general approach takes the categorization of 'linking' originally propounded by R. Mislevey, and promotes the use of social moderation in linking specific assessments to the CEF. Sounds all well and good... but there will be more! Pop back to this spot in a few months...

Click here to go to the top of the document

Educational Resources Information Center (ERIC)

This whole section represents the largest change to the first update in March 2004. The US Government closed ERIC in January 2004, thus depriving the world of one of the most significant electronic resources ever produced.

Notices relating to the closure of ERIC can be found here and here.

Some of the ERIC resources have been distributed to other web sites, but much has been lost. The only reviews that are now relevant are those for the ERIC database which can still be searched, even though no new materials will be added, and the electronic journal Practical Assessment, Research and Evaluation which started life within the ERIC Clearinghouse for Assessment and Evaluation. I am merely preserving the other reviews in this section as a reminder to readers in the future what has been destroyed through this short-sighted policy.

[The Clearninghouse on Assessment and Evaluation provides excellent services, and ERIC/AE sites are all, without exception, worth visiting. The Home Page is only available on WWW since 2002, when the gopher service was terminated. The links include a news section, a directory of essays, bibliographies and resources in assessment, a database of jobs in measurement, current testing projects, an explanation of Goals 2000, and links to other places of interest. There is also a very useful list of all discussion groups related to testing, with listserver addresses.

The ERICAE.NET page contains many of the links that the Clearinthouse Homepage presents. But this page is somewhat more recent (I think - at least I found it in September 2000). This is an excellent site to visit. Apart from the links and suggested reading, here you can find a link to the electronic journal Practical Assessment, Research and Evaluation, and links to hundreds of on-line full text articles on assessment and testing. Of course, most of this is about educational assessment in general, rather than the sub-branch of language testing. I seriously considered adding articles from these sources to my articles page, but there are just too many for me to sit down and plough through. So follow this link to see what's available.

Arguably, this should only come in the articles section, but the ERIC Digests are a useful general resource on testing and assessment that should have a link from this page as well. From this page you can select digests that you are interested in, or do a search to get a text that will provide you with a short overview of a topic in which you're interested.

The ERIC/AE news which covered new publications/reviews of books in assessment, was difficult to maintain, and was removed from their web site early in 1996, following the ERIC/AE Infoguides on Authentic Assessment, and Testing, which were removed from my links when the Resources in Language Testing pages were updated on 15 December 1995.


The most important ERIC/AE offering is the Test Locator. This service is a collaborative effort between ERIC, ETS, the Buros Institute, and Pro-Ed Publishers, and contains descriptions of over 10,000 tests, including many language tests. You can search this database, and along with the description, retrieve the address of the publisher. From late 1995 there has been a WWW interface for the test locator, but the locator itself was run from a gopher server until sometime in 2001. The entire database is now web based.

The other major contribution of ERIC is the ability to search for articles/papers/digests by keyword, author, topic (you name it) using the ERIC database search facility. Apart from the search links I include a link to a list of "known, publicly-available...sites for searching ERIC databases" from Larry Rudner. The WWW search page is attractively designed, complete with a shiny CD ROM image. The search facility allows you to specify how many "hits" you wish to retrieve. These appear on your screen as a single file through which you can scroll. The text may be printed out, or you can save it to disk for later use.

For the 1 March 1996 update of the page, Larry Rudner brought to my attention his Search Eric page, which provides a list of all the ERIC search engines known to be available. This is a very useful page, as you can actually do the searches from this site, without having to go to the location of any of the search engines listed. Rudner also explains how you can obtain copies of journal articles from the ERIC database. A very useful page indeed.]

Click here to go to the top of the document

Models, Frameworks and Scales

Any review of oral proficiency testing without reference to the American Council on the Teaching of Foreign Languages or ACTFL Guidelines is certainly lacking something - whatever you think of the ACTFL scales! I reported that the ACTFL site had vanished without a trace sometime during August 1997, but it has now reappeared in a new format (February 1998). The 1986 Guidelines are provided in full in the download section. Also available at two other sites are a complete version of the Guidelines from the Summer Institute of Linguists (SIL), and the descriptors for speaking from a company that has registered the domain name 'languagetesting.com'. These latter two were added in September 2001.

The Centre for Canadian Language Benchmarks has the following mandate (courtesy of the home page):

  • To increase the coherence and effectiveness of language instruction for newcomers to Canada, so they can become full participants in all aspects of Canadian society
  • To develop and promote the Canadian Language Benchmarks, introduce the Benchmarks to second language practitioners and encourage alignment of curriculum and program structure with the Canadian Language Benchmarks categories
  • To establish and manage a comprehensive national Canadian Language Benchmarks assessment system

The benchmarks and associated tests were developed for Citizenship and Immigration Canada by the Centre for Language Training and Assessment in Ontario. This web site informs the reader how the tests were developed, to what use they are being put, and provides many useful examples. The documentation supporting the Benchmarks is impressive, and anyone who thinks that the European Union CEFR is a useful heuristic should also look seriously at this.

A link to the Dynamic Assessment web site was added in Spring 2009. This site was brought to my attention by one of my research students. You'll find a general description of what dynamic assessment is, along with an extensive bibliography. This is really very useful for anyone working in this area, although I must confess that I can't see any references to the language testing dynamic assessment literature. There are also links to other web based resources, although I'm not sure how often these are checked and updated. Many of those that are live take you to commercially available DA materials, but a few are free resources. Perhaps the most useful part of the extensive bibliography.

Provided by Fairfax County Public Schools is the Performance Assessment for Language Students scales in speaking, writing, presentational skills, and interactive tasks. A useful list of scales that can be used in discussion groups on scale development.

Click here to go to the top of the document

Associations, Centers and Councils

Added on 10 July, 2002, is the Research Notes page. This is an excellent series of publications isssued by Cambridge ESOL to keep teachers and researchers up to date with the range of research that they undertake on their tests. The Research Notes are all available from this web page, and are also published in hard copy.

A site not to be missed is the American Psychological Association .This is a vast database of useful information, and I have found that the only way to find what you are searching for is to use the search facility which you can access from the Home Page. There are literally hundreds of documents, including the APA Standards, to be read on your screens. Not only is the information high quality, so is the presentation - especially if you access using Netscape!

The American Educational Research Association (AERA) has some very useful information available on the internet. Under Division D, which is Measurement and Research Methodology, there is information on the list AERA-D, and how to subscribe to it. Much more interesting is the section of their server which provides information on publications (such as Educational Researcher, Review of Educational Research, American Educational Research Journal, Journal of Educational and Behavioural Statistics,and Educational and Policy Analysis, and the Resources section. This latter directory allows you to search the AERA publications listed above, as well as papers presented at AERA meetings, which constitute a subset of the ERIC Current Index to Journals in Education database. You will also find here numerous Standards documents, including the Code of Fair Testing Practices. AERA also offer (probably) the only language testing related electronic journal on the internet, in the form of Rasch Measurement Transactions. This is a quarterly publication which AERA claim is the "most up to date source" of information on Rasch measurement, and to be "often months or years ahead of professional journals." Visit, and see for yourself!

The American Evaluation Association has a site which provides information on general issues in educational evaluation, with a jobs section, notices of meetings, a list of interest groups and Program Evaluation Standards. There is also information on the lists EVALTALK and EVALINFO.

A European project to support teachers through the process of exam reform was initiated by the British Council in 1998. This web site contains test specifications for the national examinations, guidelines for item writers, assessment scales and even interlocutor frames for speaking tests. Also available are full text copies of the background to the project. It is quite unusual for this type of information to be provided free of charge on a web site, and provides an invaluable insite into test development. Some commercial materials are also advertised through the site. This resource was added to the list in Spring 2005.

For 3rd February 1997 update, I came across the Homepage of CITO, the Dutch National Institute for Educational Measurement. The site contains information on the range of work undertaken by CITO in Holland, including tests designed for primary and special education, secondary, vocational, adult and industrial education. But the site promises to be more than this, as it publishes more about its research in the Measurement and Research Department. Starting off with details of its computer programs for one parameter logistic models and optimal test design, it promises to include a great deal more. This will be another one of the sites that will be well worth watching as it develops, to see if it realises its potential.

Preparing for examinations or tests? Or helping students prepare for examinations or tests? Then the Counselling and Development Centre page on preparing for tests and examinations is for you - or would have been if it hadn't evaporated into cyberspace sometime in late 2001 or early 2002. This was quite an extensive site, and offered practical tips on:

  1. Obstacles to obtaining top grades
  2. Selecting information
  3. Strategies for studying
  4. Taking multiple choice tests
  5. Getting ready for essay style examinations, and
  6. Reducing test anxiety.
If anyone needs a good idea for a new test related site, how about recreating this one?

The Center for Applied Linguistics has an extremely useful and comprehensive site. CALS also maintains a database of foreign language tests, which I came across for the 1 May 1996 update to the Resources page. The information provided says that this database was begun in 1990 in an attempt to catalogue tests in what are frequently referred to as the "less commonly taught languages". The database essentially consists of a list of languages, with hyperlinks to tests available. Information on these tests follows the same format in each case:

  • Available To:
    • Test Development Information:
    • Current users:
    • Intended Test Level:
    • Intended Test Use:
    • Skills Tested:
    • Test Author:
    • Publication Date:
    • Test Cost:
    • Test Length:
    • Test Materials:
    • Test Format:
    • Scoring Method:
    • Date of Information:
  • Description:
  • Technical Information:
  • Comments:
  • Parallel Versions in Other Languages:
  • Contact Address:

This database is certainly an excellent place to begin to look for tests you never knew existed!

The IATEFL Testing SIG page was one of the first to be added in 1998, and the link removed in September 2001. It was a straight forward page with information on the activities of the SIG and its committee, but when IATEFL reorganised its directory structure it just got lost somewhere. I spent around 30 minutes trying to find it from the IATEFL home page and failed.

Ever wondered how someone decides if a pilot has enough English to fly his plane over your house? Find out at the International Civil Aciation English Association. You may never feel entirely safe in the air again, but this is a good site to visit and provides insights into issues surrounding English for Specific Purposes teaching and testing.

The Joint Commitee on Testing Practices was formed by AERA, APA and NCME to work on testing issues relating to fairness and standards. This site lists their activities and publications, including the code of fair testing practices, and other documents related to responsible test use.

A very well designed web site and, from the lists of those attending their meetings, a very nice bunch of people. So if you're in the Midwest, this is the organization for you. Pity I'm too far away.

The "Rolls-Royce" of the sites in this category continues to be the National Center for Research on Evaluation, Standards and Student Testing (CRESST), at the University of California in Los Angeles. Not only is the quality of information excellent, the presentation is unrivalled. Both CRESST sites provide access to newsletters, research papers, technical reports (some full-text and downloadable), and CRESST products (videos, books). There is also a searchable database of tests which are classified as "Alternative Assessment." After searching through the CRESST pages, I have had additional questions/requests on three occasions, and their response to queries is also fast and efficient. To read the technical reports and newsletters you will need a PDF viewer (Adobe), but this can be d ownloaded easily enough. CREEST provides the appropriate links.

Other sites include the The National Council on Measurement in Education is somewhat more sophisticated, with the standard news and announcements page, but also a jobs section, detailing posts in educational measurement, and a database of software and shareware for test analysis. The NCME Web page was brought to my attention by Larry Rudner (ERIC) for the update of the Resources page on 1 March 1996. I visited the site again, and discovered that NCME has added to both its site a section entitled Testing Tips and Issues. The files at this site are a good read, including tips on constructing multiple choice tests, assigning letter grades to test scores, reliability issues and the like. I hope that this section is expanded in the future. I should also add that the NCME Web pages are very well designed.

An addition to this section on 1 November 1997 was the New Zealand Council for Educational Research. But they had removed the site by March 2006 and haven't left a forwarding URL. Oh dear. Another one bites the dust!The site did have some information on language testing, but quite a lot of other useful stuff on research in New Zealand. Pity it's gone.

One loss to this section in the 1 April 1996 update was the National Center on Postsecondary Teaching, Learning and Assessment, which had offered a newsletter, a diary of forthcoming educational events, and a summary of recently completed and current research projects and findings. If anyone knows where it's gone, please let me know!

The aims of the TAE SIG are to raise awareness among ELT professionals of assessment issues in the Arab world. We also want to foster professional development and research in the areas of testing, assessment and evaluation. The TAE SIG was founded in 1996 by Christine Coombe and Nancy Hubley as a forum for anyone interested in assessment issues in the Arabian Gulf. Membership has grown from a mere 10 interested testers in Al Ain to over 450 members in seven different countries.

I had the privilage of being the invited speaker at the first conference held in Al Ain by the TAE SIG in 1997, and it was great fun. (There used to be a picture of me under the CELT1 link).

Click here to go to the top of the document

Universities

The National Center for Interpretation Testing, Research and Policy at the University of Arizona "was established to meet the needs of a state and nation experiencing dramatic growth in the number of limited- and non-English speaking minorities requiring language-mediated services." Part of their work is to ensure high quality interpretation services, and they develop tests for interpreters. Link added on 30 March 2000.

Hong Kong Poytechnic University hosted the Asian Centre for Language Assessment Research from January 2000 until around November 2003. The Centre closed and the web site was removed when Liz Hamp-Lyons left HKPU and moved to the University of Melbourne.

The The Carleton Academic English Language(CAEL)Assessment is a Canadian offering to challenge TOEFL and IELTS. It is available in South America and the Far East, in addition to Canada and some other places. The site gives a description of the test, and provides sample papers and registration details.

On the 1 June, 1996, I added the link to the University of Duisburg research in language testing page, which provided information on work currently in progress there. When I checked all links in July 2009 the link was dead, and despite google searches and searches on the Duisburg web site I couldn't find it. This is a great shame, as it was extremely informative. For me, the section on the C-Test research was the most interesting. with sections entitled:

  • What is a C-Test?
  • Research into C-Tests (bibliography)
  • Relationships with other measures of language proficiency
  • Predicting difficulty of C-Tests in English and German (Readability measures)
  • Examples of C-Tests in various languages with solutions
  • Members of the project team in Duisburg
  • How to construct a C-Test
  • Glossary of test terminology

This was some really good stuff in here, helping clarify some of the mysteries of C-tests, and the glossary of statistics and test theory by Ulrich Raatz and Christine Klein-Braley was very good indeed. A quality site, much missed.

The University of Hawaii at Manoa had a Language Testing page until sometime in 2003, which gave information on current testing projects. The last time I visited this site descriptions of projects in the fields of the assessment of cross-cultural pragmatics and East Asian role plays were available. You can still go to http://www.nflrc.hawaii.edu/ and do a site search for what is going on there, but there isn't a single link.

Javeriana University hosted the Language Testing in Colombia Web Page until sometime in 2001. The site had a section for conferences, web information, resources, articles, and a who's who. I was a saddened to see this page go, as I designed it with a colleague I met in Colombia after I was invited (with Arthur Hughes) to attend the National Conference on Language Testing in 1998. The idea was to give teachers who are interested in testing their own local news and links pages, and to encourage further participation on testing projects in the country.

So what is DIALANG? Based at the University of Jyväskylä, DIALANG is a new European project which is developing diagnostic language assessment tools in 15 European languages. DIALANG is an acronym for diagnostic language assessment, and is funded by the E.U. Here is some of the blurb, but do visit the Jyväskylä pages to learn more.

  • DIALANG is a new European project which is developing diagnostic language assessment tools in 15 European languages. The 15 languages of the DIALANG project include all the official EU languages as well as Irish, Letzburgisch, Icelandic and Norwegian. The assessment materials will cover all levels from beginners to advanced; there will be separate assessments for reading, writing, listening and speaking as well as for structures and vocabulary. The project will complement traditional approaches to language assessment by exploring the use of new technologies for assessing the use of language. It will also incorporate important elements of self assessment.

Vocational Language Assessment On-line , or V-LASSO, was a collaboration of :

ACCESS taal & communicatie BVBA (Belgium), Agenzia Regionale del Lavoro della Regione Autonoma Valle d'Aosta (Italy), CITO, Dutch National Institute for Educational Measurement (Netherlands), Centro Linguistico e Audiovisivi Universitario (CLAU) Universita degli Studi di Torino (Italy), and The London Institute Centre for Languages, (UK).

Unfortunately, someone eliminated the web site in 2002. While it was there, the purpose was to provide on-line assessment in European languages for vocational purposes. The group is developing the tests, the delivery systems, and attempting to disseminate results widely. The site also stateed that the group is supported by the European Commission. Demo tests were also available. When I wrote the original review of this site, I wrote "This sounds exactly like DIALANG, but European politics are beyond the scope of this column. Thank goodness." And now it's dead and gone, I can only speculate about which other European based testing organizations or 'collaborations' helped it on its way.

Language testing has always been strong in Australia, and the University of Melbourne Department of Applied Linguistics and Language Studies' National Languages and Literacy Institute of Australia (NLLIA) has a site for its Language Testing Research Centre. The pages list the objectives of the Centre, some of its current and past projects , and a useful list of selected publications related to the Centre's Work. It used to have a second page, the Technology in Language Testing page, where a discussion on the role of technology in testing was just getting off the ground when it was discontinued.

Lancaster University set up a page for Language Testing Update on November 2, 1995. To quote from the page:

The aim of LTU is to keep language teachers, testers and researchers up to date with events in the field of language testing. The editors therefore welcome short articles (say 3 or 4 pages) about language testing research, and also news about test organisations and recent and forthcoming conferences. Articles and information can be sent to the editors by e-mail. LTU is the official newsletter of the International Language Testing Association (ILTA), and ILTA members therefore receive copies automatically.

Everyone who subscribes to LTU knows what an invaluable resource it is in keeping up with l anguage testing news and research around the world. Don't miss out!

The Language Testing Research Group at Lancaster also has a page, providing information on their members' projects, and the programme of seminars they organise. This site has grown very quickly, and the findings from many of the projects which Lancaster is currently undertaking are likely to be of great importance to the language testing community in the future. This site is well worth visiting from time to time to find out what is going on "up north" (where I originally come from - and did my PhD with Charles Alderson)!

The English Language Institute at the University of Michigan has a very well designed web page. My apologies to Mary Spaan for only adding the link on 3rd February, 1997, even though I had been told about the language testing material there a month or so earlier. From their homepage there is a link to an area on testing and certification. Information is available on the Michigan English Language Assessment Battery (MELAB), the Examination for the Certificate of Proficiency in English (ECPE) and the Examination for the Certificate of Competency in English (ECCE). With its long history of test development, the University of Michigan is worth a visit, and you can also discover which tests are commercially available.

On 2 September 1996 I added a link to the Center for Advanced Research on Language Acquisition at the University of Minnesota This site informs the reader about the Center's activities, which are fairly broad, and included the study of computer adaptive testing (CAT). The work being done on CATs is described, and there is a useful CAT FAQ.

The Test of English for Educational Purposes is now owned by the University of Reading. Originally developed for the AEB (now AQA) by Cyril Weir, this was one of the first academic English tests to be based on a thorough needs analysis. From the site you can download a practice test. Link added in the October 2002 update.

The University of Warwick has a page giving brief details of their Language Testing and Evaluation Unit, which was added on 1 January, 1997.

Click here to go to the top of the document

Archives

Some discussion lists maintain archives of debates. The TESL-L archive contained the contributions of subscribers on topics of interest, including language testing issues. Subjects which have been discussed at length include: Techniques for commenting on student papers, assessing oral proficiency, portfolio assessment, and student bias in teacher evaluation. The archive appeard not to have been updated for a long, long time, even when I first added a link to it around 1996. It passed away peacefully in 2002. Let this be a warning to people who set up web sites - to live, they need to be maintained with care.

The W3 Foreign Languages Testing Database was first mentioned on an ltest-l debate early in 1998, and it disappeared sometime in late 2000 or early 2001. The author, Rongchang Li, intended to set up a database of tests and test items that language instructors can draw upon for their own classroom uses. He said: "I believe every language instructor has his or her own small testing database. If we put them together, a big database can be created. This is indeed this Web site's intention. Any language instructor who is willing to pour his or her small database into this big database will have the right to use this database." The database was run from Filemaker Pro 4, which has a facility for entering and extracting data from the database over the web. There was a demo of the database up and running. The aim was noble, and is worth of comparison with the Goteborg experience (see on-line tests below): if you rely on others to offer their tests or items to fill out your site, chances are there will be little or no development for years. If ever. The message here is that if you decide to set up a language testing site you've got to design the content yourself and/or provide a service of some kind. No one is going to do it for you.

Click here to go to the top of the document

Ethics and Fairness in Testing

This whole section was added on 1 July 1998. For those following the conferences and literature, the number of papers being produced in this field has been growing since 1987 and no end to this is in sight. In response to this growing interest, I have added this section.

Added on 13th January 2000 is the link to the American Psychological Association's order form for the new edition of the Standards for Educational and Psychological Testing. This new 1999 edition updates the 1985 edition that I have become so familiar with. The Standards documents are quite simply essential reading and companions for anyone who works in this field. So at the risk of advertising, I'd recommend you order your copy today. In case you're unsure what the Standards are, here's a quotation from the opening paragraph of the book:

Educational and psychological testing and assessment are among the most important contributions of behavioral science to our society, providing fundamental and significant improvements over previous practices. Although not all tests are well-developed nor are all testing practices wise and beneficial, there is extensive evidence documenting the effectiveness of well-constructed tests for uses supported by validity evidence. The proper use of tests can result in wiser decisions about individuals and programs than would be the case without their use and also can provide a route to broader and more equitable access to education and employment. The improper use of tests, however, can cause considerable harm to test takers and other parties affected by test-based decisions. The intent of the Standards is to promote the sound and ethical use of tests and to provide a basis for evaluating the quality of testing practices

The Center for Equity and Excellence in Education Test Database came from George Washington University and was provided through ERIC AE. So in January 2004 this was another good web site to bite the dust because of the US Government decision to pull the plug on ERIC. It provided descriptions of 200 tests that are used most frequently with "Limited English Proficient students." You could search the database, or browse. Entries were organised by the following categories:

  • Diagnosis
  • District Evaluation
  • Identification
  • Language Dominance
  • Placement
  • Proficiency
  • Program Exit
  • Progress
  • Program Evaluation

The purpose of the databse was to help test users select appropriate tests for their needs. Now teachers will have to find some other (probably more costly) way of finding the test which best suits their students.

The Consortium for Ethics and Standards in Testing is (or rather, was) an organisation located in the School of Education at Boston College. In the March 2006 update it had disappeared, and I can't find it. This is a great pity. But as always, I leave the reviews here for historical interest and just in case the organizations appear again! Please let me know if this one does. In the meantime, here is the review of a site you can no longer visit: On the main web page is a clear indication of the aims and objectives of the page. I quote:

The Consortium for Equity in Standards and Testing (CTEST) focuses attention on how educational standards, assessments, and tests can be used more fairly.

CTEST's goals are to:

  1. Help parents, teachers, advocacy groups and other citizens understand standards, assessments, and testing;
  2. Examine the values and beliefs that underpin various proposals for standards and testing;
  3. Provide a place where parents, teachers, advocacy groups and others can discuss the positive and negative impact of testing and standards on all students;
  4. Recommend improvements in standards and testing that would identify and nurture talent, especially among racial, ethnic and linguistic minorities; and,
  5. Help parents, teachers and advocacy groups develop and disseminate information and resources about standards and testing

It does this by providing a great deal of useful information that is highly relevant to language testing and educational assessment and testing in general. In Testing in the News you can read abstracts from US papers that raise testing related issues, and Splotlight Issues has links to documents and headline news on testing issues. The documents section is especially useful, including research & journal articles along with editorials. This is an excellent site, and a must-visit for anyone interestested in equity issues.

FairTest is another site dedicated to fairness in testing. Here you will find a range of articles and documents relating to testing across disciplines and levels. To quote from the site:

The National Center for Fair & Open Testing (FairTest) is an advocacy organization working to end the abuses, misuses and flaws of standardized testing and ensure that evaluation of students and workers is fair, open, and educationally sound.

We place special emphasis on eliminating the racial, class, gender, and cultural barriers to equal opportunity posed by standardized tests, and preventing their damage to the quality of education.

The FairTest Examiner is part of the same site. The Examiner is a newspaper that deals only with the topic of fairness in testing, and many of its articles are made available in electronic format at this site. The sheer range of information available makes this site extremely important not only for issues of fairness, but just for keeping up with what is going on in testing in the United States in general.

Standards in Education is a useful list of links to a variety of Standards documents in a range of subjects. Until I discovered this page and added it in January 2001, I simply didn't know that so many of these existed. Primarily of interest to the language tester for the sheer range represented.

The National Task Force on Equity in Testing Deaf and Hard of Hearing Individuals is an excellent site, listed for the first time in the December 2001 update to this site. It is well worth copying their mission statement here in full:

Mission Statement

The mission of The National Task Force on Equity in Testing Deaf and Hard of Hearing Individuals is

  • to increase the pool of deaf and hard of hearing professionals providing service and leadership in our country today and tomorrow, and
  • to promote equity in testing deaf and hard of hearing individuals of all ages.

The Task Force includes deaf, hard of hearing, and hearing members from different professions. We work simultaneously on several fronts to carry out this mission: (1) with state and federal licensure bodies to influence policies about using tests for professional access; (2) with employers to increase understanding of how to determine the qualifications of deaf job applicants and deaf employees; (3) with college and university admissions offices to promote better understanding of reasons behind possible discrepancies between standardized admissions test scores and other aspects of applicants' academic or professional profiles to make equitable admissions decisions; (4) with colleges and universities to help deaf students become better prepared for licensing and certification tests; and (5) with testing companies to determine appropriate accommodations for deaf candidates, devise guidelines for alternative assessment procedures, and ensure equity in the development of new tests.

In carrying out this mission, the Task Force brings together persons with expertise in many different fields, so that a valuable talent pool of deaf professionals will be accessible to future students and clients in this country.

The links from this site to issues in testing disabled students is impressive, especially when it comes to news stories or breaking items. For example, the issue of 'flagging' test scores when tests are taken under conditions that are altered for a disabled test taker was 'big news' in 2000, and this site had all the issues covered, with links in to the most important web sites containing information on this fairly specific aspect of testing. Keep up the good work at Gallaudet!

Testing students with disabilities is a critical area of assessment because it throws up in very stark relief many of the issues of fairness and construct validity that are present in all testing. Take for example recent court cases over the flagging of test scores, which shows how sensitive these issues are. This site provides a good, if brief, introduction to the issue against the backdrop of the ADA legislation in the US. It has a list of official documents, followed by links to relevant articles in the press, and useful associated web sites. A useful starting point for anyone interested in this important area of testing.

Click here to go to the top of the document

Statitics

This section was added for the update on 1 May 1998. Many of my own distance learning postgraduate students had asked me where on the web to look for information on statistics, including packages that they could use. I soon discovered that there is a lot of stuff out there, but the links below are to those that I found most relevant for my students studying language testing as part of a degree in applied linguistics.

The first of these links is to JavaStat, which was brought to my attention by a posting of Fred Davidson to LTEST-L in early 1998. This is a fantastic site, providing much of what any postgraduate might need for empirical research. Instead of providing my own summary, I quote from the site itself:

"The Web pages listed here comprise a powerful, conveniently-accessible, and free multi-platform statistical software package. There are also links to online statistics books, tutorials, and related resources. The pages are located on servers all over the world, and are the result of much cleverness and hard work on the part of some very talented individuals.

These pages use various techniques to accomplish their magic:

  • Some use HTML forms to send data back to the web server, which runs CGI scripts to perform the computations and sends the results back to you as a new page. They should work on any browser that supports HTML forms.
  • Others (including all of my own JavaStats pages) use JavaScript, so each page contains within it all the programming needed to do the computations. These are flagged as (JavaScript), and should run on any JavaScript-capable browser (Netscape v2 and above, or MS Internet Explorer v3 and above).
  • Still others invoke Java applets to carry out the computations. They are flagged as (Java), and require a Java-capable browser (Netscape v3 and above, MS Internet Explorer v4 and above)."

The sheer scope of JavaStat is amazing, and the site maintainer deserves a thank you from anyone who needs to conduct empirical research

HyperStat is an introductory text book on Statistics. This text has chapters on:

  • Describing Univariate Data
  • Describing Bivariate Data
  • Introduction to Probability
  • Normal Distribution
  • Sampling Distributions
  • Point Estimation
  • Confidence Intervals
  • The Logic of Hypothesis Testing
  • Testing Hypotheses with Standard Errors
  • Power
  • Introduction to Between-Subjects ANOVA
  • Factorial Between-Subjects ANOVA
  • Within-Subjects ANOVA
  • Prediction
  • Chi Square
  • Distribution-Free Tests
  • Measuring Effect Size

Our thanks to David M. Lane for a wonderful resource.

If you want a much more advanced text, look no further than StatSoft. Once again, I can do no better than quote the text provided at this site:

"The Electronic Statistical Textbook (Statistics Homepage) offers training in the understanding and application of statistics. The material was developed at the StatSoft R&D department based on many years of teaching undergraduate and graduate statistics courses and covers a wide variety of applications, including laboratory research, business and statistics forecasting, social science statistics and survey research, data mining, engineering and quality control applications, and many others.

The Electronic Textbook begins with an overview of the relevant elementary (pivotal) concepts and continues with a more in depth exploration of specific areas of statistics, organised by 'modules,' accessible by buttons, representing classes of analytic techniques. A glossary of statistical terms and a list of references for further study are included."

The structure and organisation of this site is extremely good, and the material excellent. I was particularly impressed by the glossary, but to praise one part of this site is to suggest that the entire product is less than excellent. And what is more, StatSoft allows you to download the entire textbook onto your hard disk so that you can read it off-line. Just amazing.

From MESA Press, just what you always wanted! You can now download copies of Bigsteps and Winsteps, as well as Facets, free of charge. Needless to say these are evaluation/student versions, and put a limit on the number of data points you can enter. But for many, the evaluation copies will be just fine.

You could also get a free download from Statlets (NWP Associates). But it disappeared early in 2006. From this site you could get evaluation/student versions of a range of Java Applets to run statistical tests and create graphics. Please let me know if anyone finds it, as it had some neat programs. The free versions limited you to 100 rows and 10 columns of data. If you had more, then you needed to purchase a licence. Still, 100 x 10 isn't bad at all. There's also a manual to download, explaining how it all works, and what sort of browser you need to run the applets.

VassarStats is a set of web pages that perform statistical calculations. All done in Java. And rather amazing it is. No real explanations here. Just number crunching.

Lertap is an Excel workbook for calculating classical item statistics. Sold through ASC, it has been developed at Curtin University in Australia. Curtin also supplies a lot of on-line documentation and example files here. An independent review is available here if you are thinking of purchasing this package.

The Educational Test Analysis web site disappeared in 2006. However, the software that it offered is still available. I reproduce the review below for anyone interested.

This Review of ETS by Glenn Fulcher appeared in Language Testing Update 2005, Vol 36, 152 – 154.

ETA is a test score analysis package developed by Arthur Hughes and Anthony Woods, names that should be familiar from their book, together with Paul Fletcher, entitled Statistics in language studies which was published by CUP in 1986. An overview of the package along with ordering information is available by writing a brief note to stet2@aol.com.

This software will essentially do three things that are usually needed by students who are beginning to explore test analysis. The first is to produce classical item and test statistics for dichotomously score items, the second is to produce Rasch item (but not test) statistics for dichotomously scored items, and the third is to produce inter-rater agreement statistics for partial credit items / tasks. It is not as sophisticated or flexible as commercial software such as Iteman or Xcalibre (see www.assess.com), but as at the time of writing they are priced at $299 and $399 respectively (with a 50% discount for 10 or more licences), at just £18 (discounted to £10 for students and/or multiple copies 10) ETA is a contender for student use in a teaching context.

Data sets can be put together in Excel or specialist programs such as SPSS or SAS, or in Notepad or Wordpad, although using the latter requires more care. Cells can contain 0/1 as an incorrect/correct matrix, the selected option in a multiple choice item, or a score for an item or task. Columns on the spreadsheet refer to items, and rows to subjects, but the first three rows could be used to contain non-subject information. If the data is multiple choice the first row would contain the key, if reliability analysis is being conducted the second row would contain split half information (whether an item is to be included in the first or second half), and the third row may contain maximum score information for polytomous items.

The interface for ETA is very simple. Files are imported using the ‘files – import’ command, and these are read into the program for immediate analysis. In the ‘Analysis’ menu are four options:

• Classical item analysis
• IRT with the Rasch model
• Test reliability
• Marker reliability

In classical item analysis you get item facility and a useful histogram of the data, and a point biserial correlation between each item and test total score both with and without that item. There is also a useful table giving responses by key and distractor for distractor analysis. Total test score descriptive statistics are provided, and a histogram of score distribution.

The Rasch analysis section is the least developed of this package. What you do get is estimates of item difficulty and person ability (theta) for each test taker, with standard errors and fit statistics. What it does not provide is any item information function, data maps, or test information function information, that we have grown accustomed to seeing in packages like Xcalibre. Nevertheless, there is an option to treat items as anchors, making it relatively easy to use ETA for equating tests.

The test reliability option provides split half correlation, ANOVA estimates of variance that can be attributed to the split half and to students, Spearman Brown, KR20, Cronbach’s alpha, and the standard error of measurement for split halves and complete tests.

Finally, marker reliability is essentially an ANOVA to check differences between raters.

There are a few problems with this package. The first is that although it is simple to use, the interface is not intuitive. You need to spend an hour playing with this to see why you are doing what you’re doing. But when you’ve got it, it is very straight forward indeed. Moving away from simple appearances to practicalities, this package can handle up to 100 dichotomously scored items and 200 test takers in a single analysis. This should be adequate for most dissertation research at masters level. But for marker analysis it will only take a maximum of 10 raters and 10 tasks, the latter ceiling restricting what can be done to very simple designs.

What is disappointing for teaching purposes, however, is the lack of functionality for producing graphs, other than the histograms that we get in classical test analysis. I find it difficult, if not impossible, to explain correlation without reference to scatterplots, or the difference between means (as in rater differences in this package) without boxplots. Finally, the ‘help’ file that comes with this package is not integrated into the software, but comes as a separate .rtf and .txt file. The explanations of the basic functions of ETA are adequate, but are difficult to follow without reference to the .csv sample files. The help file and the sample files are not cross-referenced (or linked), so you have to work out which one you should be looking at for yourself. However, when you remember what this package costs, these are not major criticisms.

For use in teaching this is a real option because of price, and would only need to be supplemented by graphic displays of data that could easily be generated by other programs. ETA does all the basic things you want students of language testing to be able to do in their introductory course, at a price that is affordable for most Universities.

William Bonk has made available a number of very useful resources that can be used by students of language testing. His Resources page will be useful to all students of language testing. These Excel Spreadsheets are for the calculation of Classical Test Statistics such as alpha, KR21 and point biserial correlations. There is also a spreadsheet for distractor analysis. There are instructions for download and use also. This is a very useful free resource.

Click here to go to the top of the document

Tools Online

This section was added in March 2006. The purpose is to review tools that allow teachers and testers to develop their own tests in one way or another.

I had no idea what Hot Potatoes was until I went to an excellent presentation on the use of the software by staff from Zayed University (UAE) at TESOL Arabia in Al Ain, where I'd been invited to give a keynote talk. It was one of those presentations where the staff had really done some excellent work in designing instructional material and associated web-based assessment, and they communicated this with infectious enthusiasm. I took their handout away, and downloaded my copy almost immediately! So what is Hot Potatoes? From the University of Victoria in Canada, this is freeware (yes, free!) that includes six applications, enabling you to create interactive multiple-choice, short-answer, jumbled-sentence, crossword, matching/ordering and gap-fill exercises for the World Wide Web. Just too good to be true, right? Wrong. Go get your copy now.

The Rubric Processor is a web programme that guides you through the process of creating your own rubric (scale and band/level descriptors) for your own performance tests. The terminology here is of course US - hence the translation for those on this side of the Atlantic. Added in March 2006, this is a great resource. There is a rubric library (ready made rubrics/scales for a range of purposes), the rubric generator - a series of web pages in which you are prompted to generate your own rubrics, a Rubric Assessment Tool (a RAT - that made me giggle), and a Report Generator. And all free to download. Of course, tools like this should be used with care. It's all too easy to spend half an hour playing with the generator and think that you've got something that's ready to use in real assessments. I hope that users of this site realise that this just isn't the case. But as I said - wonderful resource that I will certainly try with my students and get them to evalute it as part of a test desgin course.

Transana is precisely what you need if you're doing discourse analysis from audio or video recorded data. This is free software, and it's simply breathtaking. Until recently it was free, but now, alas, there's a charge. I guess that this is inevitable given the quality of the software. But there is a demo available from the web site as well. I've used it to transcribe data to study domain specific language use to create rating scales, and to transcribe think-aloud protocols. Many great features, and one of the most useful DA research tools available on the internet.

Click here to go to the top of the document

Computer Based Testing

CAT Central has been created by David J. Weiss of the University of Minessota as a one-stop information point on the internet for computer adaptive testing. I have to say that this is the most up-to-date site specializing in CATS. It has extremely good bibliographies and lists of active CATs with links to the relevant web sites. It also discusses research and current issues, as well as providing a formum for discussion among researchers who are actively engaged in developing CATs. This is a very thorough site, and one which will be essential for all those working in this area. David brought my attention to his new site in December 2003, and I have added the link in March 2004.

The CAT Tutorial by Lawrence Rudner is truly amazing. For anyone who wishes to learn about how a CAT works, or for those of us who have to teach CAT, this is an excellent resource. It explains the principles and practice of computer adaptive testing, and the basics of IRT. Rudner discusses the possible advantages and limitations of CAT, and has a readable introduction to technical issues like item exposure, item pool size, and shifting parameter estimates. In the interactive section you can select the values of a, b and c paramters for your item and immediately see how these are graphed as item response and item information functions. and on top of that, you get explanations of why the items behave in the way they do. This is simply a wonderful resource, for which we are very grateful.

Click here to go to the top of the document

Tests Online

The Interactive Quiz Lis was a collection of language quizzes developed at the English Department of Northern Illinois University. There was lots of stuff that you could use in your own classroom, but unfortunately they disappeared in 2002. If anyone finds them, please let me know and I'll reinstate the link.

Bilan et orientation was a set of tests as follows:

  1. Tests de langues
  2. Tests de mathématiques pour les BTS industriels
  3. Tests de civilisation anglo-saxonne
  4. Tests d'auto-évaluation en Astrophysique

The word differentiation task is good for French speakers. Fun, but nothing highly unusual. Automatic correction makes this better than it would otherwise be. The links was added on 6th Jauary 1999. Link removed in December 2008, when I noticed that the tests had vanished!

The Computer Adaptive Test designed in Shockwave by Steve Chadwick is the first example of a CAT online as far as I am aware. The test itself is dubious, as are the level descriptors, but the presentation and range of item types is impressive for this medium. This link was added on 6th January 1999, and I assume that there will be more on-line tests of this quality before the end of the year.

Launched on 1 September, 1995, the English Language Proficiency Assessment page from Goteborg in Sweden looked like an enormous undertaking. I will quote the words of the site maintainers (as of 21 September 1995), rather than attempt my own summary of what it is they were attempting to do:

  • This site is meant to be, in due course, a real source of test materials.... We expect teachers around the world to contribute to our test bank. For the sake of supporting the difficult task of relating assessment to levels, it is our intention to collect information on international and national levels, and also to provide examples of student performance.
  • It is our intention to create a complete data-base of a wide range of tests to cover English as a Foreign Language in secondary education. It is also our intention to create a meeting-place for teachers around the world who want to discuss assessment.

How is this to be achieved? Well, the site came in a number of sections, one for each of the following topics:

  1. Proficiency Levels
    • International
    • National (ie., Sweden)
    • Samples of student work at each proficiency level
  1. Research
  2. Test Bank, and
  3. Pedagogical Forum on assessment
    • Articles on assessment
    • Whiteboard, for feedback on articles and exchange ideas

This site was under construction from the moment it was launched. The test bank contained a grammar and a vocabulary test, written for Question Mark software. To take a test you had to download "Testview", for which there were step-by-step instructions.

In my 1 September 1995 review, I wrote "The rest of the categories are empty, and if the site maintainers seriously wait for teachers and researchers to "fill them in", they may remain so. From what is there at the moment, it is clear that this site is aimed at practising teachers who are interested in practical testing issues, and require a source of language tests for use in Sweden and, possibly, elsewhere. I will certainly be keeping my eye on developments." On 2 September 1996, I reported that there had been no developments at all at this site. For the 1 December 1997 update to this page I visited the site again, and it was still exactly the same as the day it was launched. I wrote at that time: "It looks as if we can pronounce it dead. But I may pop back in another year .... Just in case!" I popped back on 26th July 1998 in preparation for 1 July update, and still no change. I popped back again for the October 2002 update. No change. I went back again at the end of February 2004, and the designers had finally put the site down. It had been entirely eradicated from the server. I think the lesson from this site and the Foreign Language Testing Database is that you can't expect others to put content in your web site!

Englishjet is a language teaching company that I had not heard of before I came across this web site in October 2002. They offer a 'level test' and then 7 'level tests' for potential customers to see where they'd fit into the organization's classes. Feedback from the first test just tells you which level you may or may not be at, and feedback from other tests give a raw numerical score with some advice like 'this test may be too difficult for you'. There is no feedback that offers any learning value.

From all the free on-line tests that I've seen come and go since 1995, the TOEIC test preparation page from Okanagan College in Canada was one of the best. Unfortunately, it went without trace either in late 2000 or early 2001 after being added on 3 October 2000. Using Shockwave, the programmers had used the director authoring package to produce some masterly practice tests. The listening practice was done particularly well, matching TOEIC type items well, and streaming effortlessly onto my machine. A great loss.

Added on July 20, you might also like to look at Dave Sperling's ESL Quiz Center These are useful little quizzes on a range of subjects from grammar to idioms, slang and words, reading, writing, current news, geography, history, science and world culture! Once again, immediate feedback, indicating correct/incorrect answers and total score. But there is no verbal feedback or explanation. Excellent fun, and useful.

These sites represent the first attempts to produce interactive language related quizzes with automatic on-line feedback, which does not require the learner to download any additional software. One thing to note with all of them is that they are very limited in the item types/tasks which students can do, and this is all to do with what is technologically possible on the web, where the multiple choice item (+ variations) and fill in the blank with a single word still dominate.

ETS started providing its own on-line practice (as of July 1996), with its own windows-based software, which you could download directly from its practice test questions page. In my first review, I wrote: "They have quite a bit of stuff up here, but what we're really interested in is the TOEFL. If this continues, these sites could even do damage to the sale of practice test books!" Someone must have noticed (or thought of this for themselves), because ETS seems to have removed this free server sometime in early 2002.

Friedman's interactive quizzes have come and gone from the internet on a regular basis. They disappeared again in early 2001. They were for learners of French, and came with a good feedback system, even though it is not specific to the particular incorrect answer of the test taker. It was expanded to include listening tasks in 1998. Pity it disappeared, as there isn't much out there on French.

French Proficiency Tests, like it says, has links to online French proficiency tests. In addition there is information on scoring / proficiency criteria in French broken down by skill (although some of these are actually generic), and lots of cultural information. It also contains lots of resources for teachers, including audio files, games, links to media. I asked two of the French teachers at my University to evaluate this site for me, and they said that while it was useful and fairly thorough, the quality of the material to which it links is variable. And one very important warning: I use a pop-up supressor, the French teachers don't. If you are going to visit this site, make sure you do have one. The site not only looks commercial, it has all the trappings of one.

For French speaking learners of English is the Cyberlanguages placement test. This is a 100 multiple choice item test that gives you feedback on your level of English, and then feeds you into the Cyberlanguages on-line courses. They also offer a few free lessons, I think. I first came across this site in January 1999.

Added on 1 January 1997 and 1 July 1998 was the test page of language tests created for the IATEFL Testing SIG by Steve Chadwick, using Shockwave. The material is still there, but now as a part of the Stuff Media site.SSteve Chadwick is a very talented designer, using Shockwave for most of his work. The quality of the test content is another matter, but for sheer inventiveness this is certainly a site to look at.

Want an on-line test of Russian? Not being able to read Russian, I have never been able to review this. But on 31 January 2002, Kate Jarvis from Newcastle University kindly sent me her comments for inclusion on this page. She wrote: "I have just tried the online Russian quiz which your page provides a link to. It is quite a long m/c quiz which tests all aspects of Russian grammar. The questions get progressively harder. The results are shown as soon as you submit the test, in two columns: your answers and the correct answers. Incorrect answers are highlighted in red and your 'approximate level' is given at the end. The results are emailed to you later. It is quite interesting as a quiz but provides no further explanation as to why answers are incorrect, and only assesses knowledge of grammar, so is a bit misleading as a level checker." After the e-mail had arrived, Kate wrote "The makers of the test emailed the result to me a few days later, with some short recommendations on how I could maintain my current language level and improve. They included my answers to the test (and the correct/incorrect responses) as an attachment." Sounds like a pretty amazing service!

The only site to be added on 1 July 1996 update of the page, Cutting Edge Call Resources, requires you to download Shockwave before you can take the tests. Like Steve Chadwick's site, I find this a pretty creative use of Shockwave. The elementary tests are visually attractive, and the more advanced tests challenging. Some of the listening tests, however, may depend on the ability to type very quickly more than comprehension of the text. Interesting research question, Jim? I have to say that the apartment game is simply brilliant. You need some additional plug-ins to do the activities, but if you have the space on your hard disk, it iw well worth the download!

Ray Meredith's LAWN (Language Assessment Web Node) disappeared from the Web during the summer of 1997, and removed on 1 September update. He had done a pretty nifty piece of work in looking at some of the scripts that are around, and set up a framework for comparing them. The scripts to run tests were available to download, and you could use four item types. One of the different things about this site was that there was an authoring page where you could design your own on-line test without having to get into html. A pity that it was removed from the net.

The On-line Exercises and Tests for the EFL Learner from Israel, added to the links on 1 July 1997 and deleted in March 2001, was a very interesting site. It was the first site to provide on-line tests that were directly linked to specific chapters in course books written by Ronald Green, presumably used within the Israeli school system. I guess that the page went down because of changes in course books or something. Another site is the Self-Study Quizzes for EFL Students, originally added on 1 December 1997. A very extensive set of links to quizzes on the web.

On 2nd September 1996 Test your Basic Skills Knowledge! disappeared from the web. It was somewhat more primitive than those reviewed above. There were sections on reading, writing and language, as well as mathematics and study skills. Questions were presented one by one, and you had to write your answer down on paper, or just keep it in your head. You were then invited to click on a link to see if you were right. The feedback explained why the correct answer was correct, and why the others were not. Perhaps it died because of its non-interactive structure. A loss on 2 June 1997 was Marko R. Riedel's Quadrilingual Language Test. It was a multiple choice translation test between English, French, German and Russian (in any combination). The feedback system detected incorrect answers, and simply gave you the correct translation. You then needed to work out where you went wrong. It is a pity this has gone off-line, as it was one of the few translation tests available.

One addition on 2nd June 1999, and removed in March 2000, was the Test yourself: English from John Moores University in Liverpool. It was quite innovative in its use of item types on the web. The feedback system was fairly primitive, providing just a score and a short text comment on your level of proficiency.

Along similar lines is the International House Interactive English Language Exercises There are some short quizes on vocabulary, grammar and idioms, followed by a quiz on flags and European countries. Nevertheless, on my first visit to this site for the 1 July 1998 update to the pages, I spent ages on the "fun" section. This is termed "online magic." You choose a card and the computer makes it disappear. Not related to language. Not a test or quiz. But I won't rest until I discover how its done. Also from International House, and added on 3 October 200 is the Level Test. Part 1 is one hundred questions in which you drag a word into its 'correct' place in a sentence. You can't skip any of these! I know. I've tried. In Part 2 you get to send a sample of wrting to a tutor.

Test your English from Colchester English Study Centre, delivered via Edunet wad added on 2 September, 1996. This is a scream. It is a traditional multiple choice test which, when completed, is e-mailed to the aforementioned Colchester English Study Centre where it is physically marked "by a CESC teacher!". The test taker is advised to print a copy of the test and mark their answers on it, so when the results are sent by e-mail ("allow up to one week") you can see where the errors were made. This test therefore currently ranks as the most primitive available on the internet. As if this waste of manpower were not enough, the questions appear to violate every guideline written about the construction of m/c items. Take items 1 - 4, for example:

  1. This man has dark
    • heads
    • head
    • hairs
    • hair
  1. and a ..........
    • beard
    • barber
    • moustaches
    • facehair
  1. He is .......... a jacket. (This full stop was in the original!)
    • wearing
    • carrying
    • having
    • holding
  1. and he is .......... a piece of paper. (As was the lower case here.)
    • wearing
    • holding
    • having
    • getting

Glad I don't have to mark these. There are 80 questions in total, and represent good material for a graduate class on the difficulties of constructing m/c items. So please, Edunet, don't take it down!

Churchill House provided a number of on-line tests for potential students to check their entrance level. These are automatic, and primiarly use the drop-down menu feature, but this is done to good effect. In the advanced section you can write a composition, and the school says it publishes the best five on its website every month. Neat idea! But it all disappeared sometime in 2008. Link removed, I'm afraid.

Translation tests used to provide, not surprisingly, some "translation tests." Well, until it disappeared from the face of the net early in the new century. It was a list of sentences to translate with answers. This is what the sites authors said about their work, preserved in this history of testing on the web:

"The following translation tests are meant to help build or strengthen argumentative skills in a given foreign language. They are based on eight pseudo-universal categories which suggest text complexity. These categories allow for speculative thinking, reasoning, refuting, making conjectures, establishing cause-effect relations. The set included on this page is based on the vocabulary that has been collected on in fairly loose manner, and is one of the three sets marked here as 'Miscellaneous'. The other sets include more strictly defined vocabulary (everyday-life objects, religion, man and interpersonal relations) or oscilate round some specific linguistic phenomena (metaphors, idioms, grammatical tenses). The text concluding this section presents briefly possibilities that the familiarity with the specified eight categories creates".

Click here to go to the top of the document

Private/commercial organizations

Assessment Systems Corporation (ASC) set up a gopher site in February 1995, and followed this with a Web site in January 1996. The launches were published in their annual catalogues, which those of us who have bought software packages from them get on a regular basis. ASC is the producer of the FastTEST software package, and a wide range of other test analysis products. These are among the best on the market. One of the advantages of the ASC site is that you can actually try out some of the software before deciding to purchase, through the program demonstration software file. General information on all ASC programs and computer adaptive testing is available, and there is a file listing books on testing which can be purchased directly from ASC.

Until March 200 you could download Scrutiny!, produced by Advanced Psychometrics Inc. Advanced Psychometrics Inc. (API) "is [or was, as it's web site has now disappeared] a company which creates software for educational and psychological testing and provides consultation in test and scale development, test centre control, and data management."

Another company which produces Computer based assessment systems is Question Mark Computing Ltd., located in the UK. The Question Mark software has been around for some time, and many language testers are familiar with it. Question Mark have now produced a Web version of their testing software, which could allow language testers to use the Web as a delivery system. At the Question Mark site you can see a demonstration of how this system works, and check out the range of links they have to web delivered tests and surveys. The Perl scripts you have to purchase to set up tests on the Web are pretty expensive, but could be worth it for a large organization which needs to deliver (non-secure) progress or placement tests to large numbers of students. Whatever the cost or issues surrounding test security, Question Mark is showing that Web delivery is possible, and not to pursue this use of the Internet would be foolish. I therefore recommend a visit to the Question Mark site to see what they're doing.

Ed Tech had a web site until early 2001, advertising the tool QuizMaker which was designed to deliver tests over the web. Unlike QuestionMark software, this was offered free of charge to anyone working in education. Out of the University of Hawaii, the page disappeared sometime early in 2001. Perhaps business wasn't that good.

If you want to prepare for the TOEFL on-line, then check out the TOEFL Den, which I added to the links on 1 December, 1997. This site claims to be:

  • ...the first and only personal TOEFL training program on the Internet. This unique Internet service creates a personal study program to show students exactly what to study to achieve a high TOEFL score.

If you wish to listen to over 16 hours of audio (plus double the time downloading!) and answer over 7000 TOEFL questions with the virtual trainers "Max and Meg", then don't forget to ask what percentage of the fee is paid into the account to take care of the psychiatric bills.

Knowledge 3000 is yet another company offering preparation for the TOEFL, and was the first commercial operation geared specifically for the CBT. When I added this site on 6th January 1999, the following quotation was placed at the top of the web page:

  • "Students with a high level of computer familiarity have significantly higher TOEFL® scores than those examinees with a low level of computer familiarity".
    (Computer Familiarity & Test Performance, 1998, p. 8, Educational Testing Service, New York)

The message is obviously that you can get a higher TOEFL score through the practice the company provides using computer based materials. The site also offers some sample questions, but really wishes you to purchase the "Cracking the TOEFL" package.

The Princeton Review provides advice on tests, how to pass them, and what grades you would need for entry to particular careers and courses in the US. There is also a Japanese edition, linked into this site on 1 July 1998, which offers free online TOEFL and TOEIC style tests, news on courses and a course locator. Most of the text is in Japanese, and you also have to have a password to get access to some parts of the site.

Professor TOEFL's Fun Page from Encomium Publications - which suddenly disappeared from the internet in March 2000, was absolutely awful. This is a great example of how this document is rapidly turning into a hideous monster. Goodness only knows when I'm going to get around to writing it out again. But at least this serves a historical purpose, I guess. Clearly commercial in orientation, TOEFL takers were given the opportunity to purchase practice materials, as well as ask questions of Professor TOEFL, like this one from Cristiane in Brazil (which I'm glad I copied out now): "Would you please show me an example of a reduced adverb clause and a reduced adjective clause and explain to me what the difference is?" If you can ask a question like that correctly, I don't think you need to worry about your TOEFL score!

Testwise is a commercial site offering TOEFL Video products. However, they do have a 'free' review of the TOEFL, with example questions and suggested answers. Non- interactive in nature, it is their attempt to show students (and potential clients) what sort of thing it is they are selling. Fairly run-of-the-mill stuff.

Edutest provide an on-line testing service that you have to pay for in advance. The basic idea is that parents will wish to pay to have their child's "specific areas of strengths and needs" evaluated by a range of "language arts" and other tests. The site says that all the tests have been constructed by sampling the Virginia Standards of Learning, which are also available on-line. The claim is therefore that they offer criterion referenced testing for diagnostic purposes - but they don't actually say this. This site, added to the links on 1 July 1997, will go down as the fist pay-for-use on-line testing. But I am sure that it won't be the last. I wrote that sentence quite a long time ago! I lost track of this one in 2007 - so it didn't do too badly!

Another commercial site, added on 3rd February, 1997, was Encomium, which published the TOEFL MENTOR. The site disappeared sometime in 2002. If anyone finds it, let me know! What follows was the original review: "This is a CD-ROM preparation system for the TOEFL, covering all parts of the examination. You can download Mac or PC demons of the software from the web page, and try it out. So, you can see the page, and you can also read the review by Tom Robb. If your business is TOEFL preparation courses and you have access to those costly multimedia teaching labs., then this just may be what you're looking for."

International House's (IH) Web site had a useful page on British English Language Examinations, but in March 2001 I discovered that it had disappeared and in its place was a commercial site advertising language courses. That isn't playing fair. So others, be warned - I will check sites periodically, remove links, and put up comments like this to embarrass the offenders!

A similar site to International House was that at the Vancouver Language Institute, which disappeared from the Web sometime in March or early April 1997.

Information on the Test of English for International Communication (TOEIC), was also offered by ICI International Communications, but this disappeared sometime in December 2003. The site had information on the use of TOEIC in companies and how you can take the test. ICI claimed to be able to help learners in preparing for the test, helping with corporate language planning, and setting minimum scores for certain company positions.

The Friends of TOEIC Page which I discovered in February 1997 was quite amazing, even though I couldn't read it. Mostly in Japanese, the site had book reviews (presumably TOEIC practice test books), study guides, a careers section, and a place where the Friends could post messages to each other. In September 1998, however, the site changed into the official TOEIC web site of Japan. It's still all in Japanese (anyone care to send me some translations?) but it now looks much more sober. The usual descriptions, information and statistics, for the most part. No longer off-the-wall. The LACS Centre (we aren't told what this means - or at least we weren't when I first visited the site in April 1999 and added the link on 1 May 1999) is another company that offers to sell you test delivery software, or design tests to meet your own needs. You can download samples, including sample software, from the web page. There are also some on-line tests here. Pleasant site. But what always worries me when I discover another one of these commercial sites is that there never seems to be any awareness of wider language testing issues, the qualities of good tests, or professional test design. I guess that's just one of the problems of the push towards computer based tests, that a lot of test design is going to be technology driven, however well meaning the people who design the tests may be. This kind of approach should be contrasted with Ordinate and Second Language Testing Inc., that have a very strong research base.

The RAND Institute for Education and Training moved all its testing and assessment material to a World Wide Web site from their old gopher site in June, 1996, and the new link was established from the Resources page on 1 July. Although the quality of presentation has (obviously) improved, the quality of the information remains poor, unless you are just keen to know the range of RAND involvement in education and assessment.

Score it Now is a commercial service from ETS launched in 2002, using e-rater technology. Once you purchase your ID and password, you can write essays on TOEFL prompts and submit them to e-rater, which will give you a score. There is also a section that allows you to look at sample essays with scores, and review TOEFL type prompts, as well as providing tips on how to improve your writing skills.

Sylvan Psychometric was also added on 6th January 1999. They changed their name after a merger to Thomson Prometric. This is certainly a site worth visiting if you are interested in Computer Based Testing, as Prometric is one of the world's leader in providing facilities for CBT to be carried out. The CBT TOEFL, launched in 1998, is delivered mainly by Prometric. Here is a sample of their own description of their role in CBT:

  • Paper-and-pencil testing has long been the primary means of assessing knowledge and abilities. Technology-based assessment (TBA) has radically changed that, making the testing process easier and the results more reliable.
    TBA enables certification/credentials exams to be administered on request, scored immediately and updated readily. Given all this, it is not surprising that the demand for it has increased substantially in business, education and government. Converting to TBA will inform people that your organization is current with a trend that is here to stay, enhancing both your position and ability to compete in the marketplace.
    We operate more than 1,500 sites in over 80 countries.
    Thomson Prometric is in a unique position to provide TBA for your organization. We are the largest division of Sylvan Learning Systems, Inc. and the world’s leading provider of secure, technology-based testing. We have over 1,500 sites in more than 80 countries, including regional support centers in Africa, Australia, Europe, Latin America/Caribbean, and East and Southeast Asia. Regardless of location, all our testing centers meet the same high standards of security, comfort and professionalism.
    We also offer varying levels of partnership opportunities to third-party training centers, including sales and marketing support. This enables them to provide the complete spectrum of services, training through testing, to candidates. In short, we are committed to the TBA field in every possible way.

Also a well designed site with good graphics.

TestServe was a company that I came across late in 1998, and added the link in the 6th January 1999 update. Then it disappeared, and I removed the link on 30 March 2000. Like Sylvan it is, or was, a test delivery service, but only operated on the internet. You could create your own test on-line, and have it delivered through TestServe servers. This was the first example of a commercial service for delivering tests over the internet. I wonder what happened to them?

ETS Technologies was added in the January 2001 update, and removed in the March 2006 update. ETS is forever restructuring, and bringing this subsidiary to an end is probably just part of the regular changes that occur. However, I leave the review below for historical interest:
And the site is high tech, with the use of Flash on the entry page - the first language testing site that uses this Macromedia product. We are told that:

ETS Technologies, Inc. is a subsidiary of Educational Testing Service. Our mission is to identify, develop, and deploy innovative technologies in support of online learning and assessment applications. Our commitment is to expanding and improving educational opportunities by harnessing proven technologies in the service of learning.

When I added this site, ETS Technologies were advertising Criterion that uses e-rater to score essays, and c-rater that scores short response answers. For further information on e-rater see the e-rater research page from ETS. Undoubtedly these new technologies will be thoroughly tested over the coming years to see if computer scoring of extended language production is indeed possible without the intervention of human raters.

The Buros Institute of Mental Measurement has an attractive Web site, which has a news section, articles, links to other related sites, and e-mail addresses for staff. The articles are a good read, and there is a FAQ on Psychological Tests. Under the links section, you can get into the Buros Test Review Locator.

TESTLINK is a new web page expanding the ETS test database. Added in March 2004, this page claims to provide searchable access to over 2000 tests. A search will provide you with a list of tests that matches the search term you enter, an abstract that explains the main purpose of the test, and the contact details of the producer. A very useful addition to the other test locator facilities currently available.

The International Baccalaureate had a Web site, which provided information on the entire curriculum, including languages, but this was taken off line sometime in early May, 1996. Furthermore, no forwarding address was left, thus leaving me with a useless link for perhaps two weeks. This is really irritating. If sites are taken off line, it is only good manners for a page to be left for a couple of months saying where the site has gone to, or simply informing the visitor that it is no longer available.

Another commercial testing service, established in 1996, is Ordinate. This was taken over by Harcourt Assessment in 2005.

This came from the original Ordinate Homepage, and is worth preserving:

  • “Ordinate Corporation brings years of research in automated language assessment to educators and human resource professionals. ORDINATE offers the PhonePass test , a testing service that is as close and easy to use as your own telephone. ORDINATE promises efficient, reliable, and affordable assessment of spoken English skills, with fast delivery of results.”

PhonePass is a ten minute test that is conducted on the telephone. The test designers claim that it can measure speaking, listening and reading skills. It is also claimed to measure “skills that support conversational proficiency in American English….” Moving in the direction of specifics, we are told:

  • “Conversational proficiency is the ability to understand and speak intelligibly at a conversational pace on everyday topics. Conversational proficiency incorporates fluency, pronunciation and alacrity in speaking, reciting and reading aloud, as well as productive control of common vocabulary and basic sentence structure in repeating sentences and answering short questions.”

As well as the test description, cost and the other types of information that you find on other web sites that advertise tests, Ordinate provides on-line reports into the reliability and validity of the test. This is to be applauded, and other larger testing organisations (ALTE members immediately come to mind) should take note that smaller companies like Ordinate are way ahead of the game when it comes to providing such crucial information.

All of the crucial information has been retained on the Harcourt Pages.

You can also take a sample test from the site, which is a very useful innovation.

One of the casualties from this section in 2 September 1996 update was or the Northwest Regional Educational Laboratory (NWREL). It contained bibliographies relevant to Portfolio Assessment and testing reading, but suddenly disappeared without trace! However, I found it again for the 13th January 2000 update! It still contains a lot of information on the assessment of writing, but now also contains information on oral assessment and program evaluation. Together with a searchable library, this is an interesting site to visit. I just hope it sticks around this time.

Click here to go to the top of the document

Conclusion

We have briefly reviewed a number of sites which are currently providing language testing and language testing related information on the internet. Although there is variation in the quality and amount of information on language testing related themes currently available, it is hoped that information providers will maintain and update their pages on a regular basis, and new information providers will come along.

If you are aware of any additional sites of importance to language testing, please let me know, so that the Resources in Language Testing Page may be updated. Do also include a link to the Language Testing Resources Site from your own web page.

Click here to go to the top of the document