Interview with Douglas M. Dunn, Ph.D., Coauthor of the Peabody Picture Vocabulary Test Fourth Edition
Share:
Schreiber:: Doug, thank you for taking the time this afternoon to tell us about the 4th edition of the Peabody Picture Vocabulary Test. Our readers are interested in knowing about the revision of this widely used test. First though, you have an interesting background as co-author of this test. Would
Schreiber:: Doug, thank you for taking the time this afternoon to tell us about the 4th edition of the Peabody Picture Vocabulary Test. Our readers are interested in knowing about the revision of this widely used test. First though, you have an interesting background as co-author of this test. Would you tell us about it?
Dunn: I am the only child of Lloyd and Leota Dunn. Dad was the founding author of the PPVT (among other tests and publications). And that product was developed over 50 years ago. Throughout the development of the PPVT, Mother and I were involved. She was a school teacher, as was dad at the beginning, and a high school principal too. So I was involved in the development of the test, but really it was only towards the end that I got actively involved with this new version of the PPVT-4.
Schreiber:: Your background is not in education or speech-language pathology?
Dunn: My background is in statistics. I have a Ph.D. in statistics and business from the University of Michigan. I worked at Bell Laboratories Research in Murray Hill back when AT&T was the telephone company and Bell Labs Murray Hill was the research arm. I had a nice career there, moved into management, into the general departments, and worked for AT&T as a corporate officer. When AT&T changed, I decided it was time for me to change as well. So I went back into the academic world and became the Dean of what was then called the Graduate School of Industrial Administration, now called the Tepper School of Business at Carnegie Mellon University. And I had a nice opportunity there; stayed for six years, and have recently retired to Las Vegas.
Schreiber:: I would assume then that your involvement in the PPVT tapped into your business and statistical background?
Dunn: My involvement was that; but you know, in small families, the involvement is anything from being the guinea pig who is tested on a new set of words trying to figure out if the set of words works, or if the decoys work, to "Doug, what's the actual meaning of this word?"--particularly if it was a technical term--a statistical or mathematical term. I was involved in that regard. And you know another editor never hurts--a new set of eyes. And so I would look over some of the plates and make sure that everything worked well, that decoys weren't too strong, that other issues hadn't been overlooked.
Schreiber:: So your father might have been the author but the "family" really played a role in the creation of the PPVT.
Schreiber:: Our readers know the Peabody Picture Vocabulary Test. It's widely used and it's been around for a long time. They know it's a norm-referenced measurement of an individual's receptive vocabulary and now it's in its 4th version. What they may not be aware of is why a measure of something so narrow (receptive vocabulary), would have implications on a lot of fronts for individuals. Could you tell us why a measure of receptive vocabulary has so much power?
Dunn: Vocabulary tends to be a key ability that we all recognize. When one looks at the value of vocabulary, one probably begins with the ability to read--to be literate. And one can hardly be literate if he or she doesn't have a reasonable vocabulary and a vocabulary that can used to expand and grow as specialized interests expand and grow. Simply to be able to read, to understand and to comprehend means that you understand the vocabulary. So, receptive vocabulary is important. Further, we often make special use of oral communication. We're doing it right here today in this interview. And of course the ability to speak--to be understood--is very important. And to understand what is being said is important. So vocabulary is a key to developing knowledge and transmitting that knowledge to others. Again I'll emphasize the importance of oral communication and the ability to understand what is said orally. But the ability to write and communicate in verbal form is just as important. So vocabulary is a key to our ability to function in a knowledgeable society.
Schreiber:: Receptive vocabulary is an indicator of language competence and that language competence also supports reading comprehension and consequently our writing competence as you mentioned. So what's new in the PPVT-4? Why the revision?
Dunn: Some magic things happen in our society. I'm particularly attuned with what's happening in technology. But with technology comes all sorts of advances in vocabulary. People will learn different words at different times. Certain words will simply drop from the vocabulary and new words will come on. And we've been particularly concerned that the PPVT be modern. One of Dad's most critical imperatives was to do things right. And by that I mean just truly to have excellence. And part of being excellent is having a test that doesn't have biases. So we want to be careful as the population changes, as people migrate and move, as different ethnic backgrounds become more prevalent, we have to be even more careful that we're not using words that are biased in any sense: gender bias, racial bias, geographic bias, and ethnic bias--any of the biases. Further, we've always wanted this test to be something that was gripping, that people wanted to do, and that was exciting. People like to be challenged with "show me a particular word." And in the new multimedia world it seems that color is an important indicator of being exciting, of being interesting. In this new version--the PPVT-4--we've been very careful with regard to the stimulus words and to the decoys. We've checked the biases very carefully. And we've introduced color, and the color certainly makes it more interesting, particularly for the younger subjects. Subjects are engaged in the color of the illustrations. It's fun.
By the way, we had to be very careful with color-blindness because that affects people, particularly males. But we were able to address that issue, and it turned out not to be a major challenge. The exciting parts of the revision were first, we made sure that the words were current, up-to-date, and representative of the modern vocabulary of subjects across the relevant age-ranges; second, that any biases that had crept in were removed; and third, we introduced color and increased the size of the illustrations a little bit to make them more gripping and a more pleasurable experience both for the subject and for the examiner.
Schreiber:: So when you addressed bias, not only did you look at the test vocabulary for bias, you looked at the balance of diversity in the art, the illustrations.
Dunn: Yes we did and we were very careful to make sure that no real or artificial stereotypes had been included. We were careful with regard to the different individuals who are in professional, or athletic, or any types of endeavors--we looked for balance across all of those dimensions. An interesting bias that also begins to creep in if you're not careful is an international bias. One of the interesting examples we found was a picture of an explosion that included a very famous bridge in England, which our subjects from across the pond, and maybe the immigrants to this country, felt was really inappropriate. Again we had to consider the international world; having currency that's only green doesn't work anymore. Currency comes in different colors.
Schreiber:: I know the development of the PPVT-4 took about five years; you had a much larger sample size for this version and many more reviewers involved, correct?
Dunn: That's all true. Again, sample size is important because you'd like to keep the variance down and you'd like to understand how well your test performs. It's also important because it allows you to have a broad cross-section of the population, and again issues such as regional and geographic bias within the United States are important to address, and a large sample size helps there. But I must tell you that what I was most pleased with is that we did this in a sequential fashion. We thought we had good ideas, and like most ideas, you have many more than you can actually use. So we did initial trials and on the basis of those initial trials, we needed to do a second set of trials and finally the standardization. So this whole process was intensive in terms of data collection. In terms of review at the beginning, the issues of cultural bias were very important to us so we found people who had different cultural backgrounds who were able to review our proposed changes and give us insights. And of course, data are our friends as I like to say, so if you know the demographic characteristics of the subjects taking the test or particularly taking the preliminary trials of these new items, you can begin to see if any biases are creeping into the tests.
Schreiber:: And the other interesting feature of the PPVT-4 is that it is co-normed with the second edition of the Expressive Vocabulary Test (EVT-2).
Dunn: Well the EVT is also published by AGS/Pearson, and they were going through a revision and many people who like to use the PPVT like to also use the EVT to understand both the expressive side of one's vocabulary as well as the receptive side. So, often the tests are used in parallel. And we wanted to be careful to make sure that there was no learning going on. So one of the things that AGS did very well, and Kathleen Williams [author of the EVT] and others working on the EVT, was to make sure that there was really minimal overlap between the two tests and that there would not be learning from one test to the other.
Schreiber:: And the revision now addresses a younger age I believe; it measures receptive vocabulary as low as 2
Dunn: I am the only child of Lloyd and Leota Dunn. Dad was the founding author of the PPVT (among other tests and publications). And that product was developed over 50 years ago. Throughout the development of the PPVT, Mother and I were involved. She was a school teacher, as was dad at the beginning, and a high school principal too. So I was involved in the development of the test, but really it was only towards the end that I got actively involved with this new version of the PPVT-4.
Schreiber:: Your background is not in education or speech-language pathology?
Dunn: My background is in statistics. I have a Ph.D. in statistics and business from the University of Michigan. I worked at Bell Laboratories Research in Murray Hill back when AT&T was the telephone company and Bell Labs Murray Hill was the research arm. I had a nice career there, moved into management, into the general departments, and worked for AT&T as a corporate officer. When AT&T changed, I decided it was time for me to change as well. So I went back into the academic world and became the Dean of what was then called the Graduate School of Industrial Administration, now called the Tepper School of Business at Carnegie Mellon University. And I had a nice opportunity there; stayed for six years, and have recently retired to Las Vegas.
Schreiber:: I would assume then that your involvement in the PPVT tapped into your business and statistical background?
Dunn: My involvement was that; but you know, in small families, the involvement is anything from being the guinea pig who is tested on a new set of words trying to figure out if the set of words works, or if the decoys work, to "Doug, what's the actual meaning of this word?"--particularly if it was a technical term--a statistical or mathematical term. I was involved in that regard. And you know another editor never hurts--a new set of eyes. And so I would look over some of the plates and make sure that everything worked well, that decoys weren't too strong, that other issues hadn't been overlooked.
Schreiber:: So your father might have been the author but the "family" really played a role in the creation of the PPVT.
Schreiber:: Our readers know the Peabody Picture Vocabulary Test. It's widely used and it's been around for a long time. They know it's a norm-referenced measurement of an individual's receptive vocabulary and now it's in its 4th version. What they may not be aware of is why a measure of something so narrow (receptive vocabulary), would have implications on a lot of fronts for individuals. Could you tell us why a measure of receptive vocabulary has so much power?
Dunn: Vocabulary tends to be a key ability that we all recognize. When one looks at the value of vocabulary, one probably begins with the ability to read--to be literate. And one can hardly be literate if he or she doesn't have a reasonable vocabulary and a vocabulary that can used to expand and grow as specialized interests expand and grow. Simply to be able to read, to understand and to comprehend means that you understand the vocabulary. So, receptive vocabulary is important. Further, we often make special use of oral communication. We're doing it right here today in this interview. And of course the ability to speak--to be understood--is very important. And to understand what is being said is important. So vocabulary is a key to developing knowledge and transmitting that knowledge to others. Again I'll emphasize the importance of oral communication and the ability to understand what is said orally. But the ability to write and communicate in verbal form is just as important. So vocabulary is a key to our ability to function in a knowledgeable society.
Schreiber:: Receptive vocabulary is an indicator of language competence and that language competence also supports reading comprehension and consequently our writing competence as you mentioned. So what's new in the PPVT-4? Why the revision?
Dunn: Some magic things happen in our society. I'm particularly attuned with what's happening in technology. But with technology comes all sorts of advances in vocabulary. People will learn different words at different times. Certain words will simply drop from the vocabulary and new words will come on. And we've been particularly concerned that the PPVT be modern. One of Dad's most critical imperatives was to do things right. And by that I mean just truly to have excellence. And part of being excellent is having a test that doesn't have biases. So we want to be careful as the population changes, as people migrate and move, as different ethnic backgrounds become more prevalent, we have to be even more careful that we're not using words that are biased in any sense: gender bias, racial bias, geographic bias, and ethnic bias--any of the biases. Further, we've always wanted this test to be something that was gripping, that people wanted to do, and that was exciting. People like to be challenged with "show me a particular word." And in the new multimedia world it seems that color is an important indicator of being exciting, of being interesting. In this new version--the PPVT-4--we've been very careful with regard to the stimulus words and to the decoys. We've checked the biases very carefully. And we've introduced color, and the color certainly makes it more interesting, particularly for the younger subjects. Subjects are engaged in the color of the illustrations. It's fun.
By the way, we had to be very careful with color-blindness because that affects people, particularly males. But we were able to address that issue, and it turned out not to be a major challenge. The exciting parts of the revision were first, we made sure that the words were current, up-to-date, and representative of the modern vocabulary of subjects across the relevant age-ranges; second, that any biases that had crept in were removed; and third, we introduced color and increased the size of the illustrations a little bit to make them more gripping and a more pleasurable experience both for the subject and for the examiner.
Schreiber:: So when you addressed bias, not only did you look at the test vocabulary for bias, you looked at the balance of diversity in the art, the illustrations.
Dunn: Yes we did and we were very careful to make sure that no real or artificial stereotypes had been included. We were careful with regard to the different individuals who are in professional, or athletic, or any types of endeavors--we looked for balance across all of those dimensions. An interesting bias that also begins to creep in if you're not careful is an international bias. One of the interesting examples we found was a picture of an explosion that included a very famous bridge in England, which our subjects from across the pond, and maybe the immigrants to this country, felt was really inappropriate. Again we had to consider the international world; having currency that's only green doesn't work anymore. Currency comes in different colors.
Schreiber:: I know the development of the PPVT-4 took about five years; you had a much larger sample size for this version and many more reviewers involved, correct?
Dunn: That's all true. Again, sample size is important because you'd like to keep the variance down and you'd like to understand how well your test performs. It's also important because it allows you to have a broad cross-section of the population, and again issues such as regional and geographic bias within the United States are important to address, and a large sample size helps there. But I must tell you that what I was most pleased with is that we did this in a sequential fashion. We thought we had good ideas, and like most ideas, you have many more than you can actually use. So we did initial trials and on the basis of those initial trials, we needed to do a second set of trials and finally the standardization. So this whole process was intensive in terms of data collection. In terms of review at the beginning, the issues of cultural bias were very important to us so we found people who had different cultural backgrounds who were able to review our proposed changes and give us insights. And of course, data are our friends as I like to say, so if you know the demographic characteristics of the subjects taking the test or particularly taking the preliminary trials of these new items, you can begin to see if any biases are creeping into the tests.
Schreiber:: And the other interesting feature of the PPVT-4 is that it is co-normed with the second edition of the Expressive Vocabulary Test (EVT-2).
Dunn: Well the EVT is also published by AGS/Pearson, and they were going through a revision and many people who like to use the PPVT like to also use the EVT to understand both the expressive side of one's vocabulary as well as the receptive side. So, often the tests are used in parallel. And we wanted to be careful to make sure that there was no learning going on. So one of the things that AGS did very well, and Kathleen Williams [author of the EVT] and others working on the EVT, was to make sure that there was really minimal overlap between the two tests and that there would not be learning from one test to the other.
Schreiber:: And the revision now addresses a younger age I believe; it measures receptive vocabulary as low as 2