ACS standardized exams versus conventional papers at a British university

B. G. Gowenlock, D. M. McIntosh, A. W. Mackaill

Research output: Contribution to journalArticle

Abstract

The relatively high correlation coefficients obtained between the Moray House Tests and the Co-operative Examinations are interesting, and it was this which prompted the investigation of performance on recall and higher processes items. The interpretation of correlation coefficients must, however, be approached with caution. Decisions on the nature of a good correlation must be made with other considerations, such as the nature of the sample, kept in mind. Correlations found in educational research tend to be rather smaller than those calculated in other areas of research. The reason for this is that, as the variables examined are usually complex in nature, it is unreasonable to expect high correlations among them. The most important decision which has to be made is whether a perceptible correlation exists. In this context a correlation of 0.3 can be considered interesting and one of 0.5 to be of some importance. Correlation between two tests does not necessarily indicate a measure of causation. The comparison of means for both 1968-69 and 1969-70 indicates that students of higher than average verbal ability scored better on higher processes than on recall items whereas students of less than average verbal ability scored equally on both types. Students of higher than average verbal ability also scored significantly higher on both types of items than students of less than average verbal ability. It has long been thought that objective tests offer an opportunity to students of lower verbal ability to score better marks. The evidence of this present research indicates that, while this might very well be the case, it is clear that objective tests do not have a deleterious effect on the performance of the verbally more able students. The lack of significant difference in the performances of students from different home backgrounds was surprising. Early investigations showed that the children of routine workers tend, on the whole, to be less able verbally, and that objective testing might well suit them in examinations. The lack of significance in this investigation might well be explained by a "masking effect" of individuals in the middle range-say socio-economic groups III to VI. Examination of groups I and II compared with group VII might well show significant differences but was not possible in this case as the number of students was too small. It is worth considering that if, as suspected, conventional examinations place undue emphasis on the ability to recall plain facts, there is not much point in replacing them with objective tests if all that would then be measured was verbal ability or intelligence quotient.

Original languageEnglish
Pages (from-to)139-140
Number of pages2
JournalJournal of Chemical Education
Volume50
Issue number2
Publication statusPublished - 1973

Fingerprint

university
ability
student
examination
intelligence quotient
performance
say
Group
lack
educational research
worker
interpretation
evidence
economics

Cite this

Gowenlock, B. G., McIntosh, D. M., & Mackaill, A. W. (1973). ACS standardized exams versus conventional papers at a British university. Journal of Chemical Education, 50(2), 139-140.
Gowenlock, B. G. ; McIntosh, D. M. ; Mackaill, A. W. / ACS standardized exams versus conventional papers at a British university. In: Journal of Chemical Education. 1973 ; Vol. 50, No. 2. pp. 139-140.
@article{da1deb824afd4882ae2a753814002372,
title = "ACS standardized exams versus conventional papers at a British university",
abstract = "The relatively high correlation coefficients obtained between the Moray House Tests and the Co-operative Examinations are interesting, and it was this which prompted the investigation of performance on recall and higher processes items. The interpretation of correlation coefficients must, however, be approached with caution. Decisions on the nature of a good correlation must be made with other considerations, such as the nature of the sample, kept in mind. Correlations found in educational research tend to be rather smaller than those calculated in other areas of research. The reason for this is that, as the variables examined are usually complex in nature, it is unreasonable to expect high correlations among them. The most important decision which has to be made is whether a perceptible correlation exists. In this context a correlation of 0.3 can be considered interesting and one of 0.5 to be of some importance. Correlation between two tests does not necessarily indicate a measure of causation. The comparison of means for both 1968-69 and 1969-70 indicates that students of higher than average verbal ability scored better on higher processes than on recall items whereas students of less than average verbal ability scored equally on both types. Students of higher than average verbal ability also scored significantly higher on both types of items than students of less than average verbal ability. It has long been thought that objective tests offer an opportunity to students of lower verbal ability to score better marks. The evidence of this present research indicates that, while this might very well be the case, it is clear that objective tests do not have a deleterious effect on the performance of the verbally more able students. The lack of significant difference in the performances of students from different home backgrounds was surprising. Early investigations showed that the children of routine workers tend, on the whole, to be less able verbally, and that objective testing might well suit them in examinations. The lack of significance in this investigation might well be explained by a {"}masking effect{"} of individuals in the middle range-say socio-economic groups III to VI. Examination of groups I and II compared with group VII might well show significant differences but was not possible in this case as the number of students was too small. It is worth considering that if, as suspected, conventional examinations place undue emphasis on the ability to recall plain facts, there is not much point in replacing them with objective tests if all that would then be measured was verbal ability or intelligence quotient.",
author = "Gowenlock, {B. G.} and McIntosh, {D. M.} and Mackaill, {A. W.}",
year = "1973",
language = "English",
volume = "50",
pages = "139--140",
journal = "Journal of Chemical Education",
issn = "0021-9584",
publisher = "American Chemical Society",
number = "2",

}

Gowenlock, BG, McIntosh, DM & Mackaill, AW 1973, 'ACS standardized exams versus conventional papers at a British university', Journal of Chemical Education, vol. 50, no. 2, pp. 139-140.

ACS standardized exams versus conventional papers at a British university. / Gowenlock, B. G.; McIntosh, D. M.; Mackaill, A. W.

In: Journal of Chemical Education, Vol. 50, No. 2, 1973, p. 139-140.

Research output: Contribution to journalArticle

TY - JOUR

T1 - ACS standardized exams versus conventional papers at a British university

AU - Gowenlock, B. G.

AU - McIntosh, D. M.

AU - Mackaill, A. W.

PY - 1973

Y1 - 1973

N2 - The relatively high correlation coefficients obtained between the Moray House Tests and the Co-operative Examinations are interesting, and it was this which prompted the investigation of performance on recall and higher processes items. The interpretation of correlation coefficients must, however, be approached with caution. Decisions on the nature of a good correlation must be made with other considerations, such as the nature of the sample, kept in mind. Correlations found in educational research tend to be rather smaller than those calculated in other areas of research. The reason for this is that, as the variables examined are usually complex in nature, it is unreasonable to expect high correlations among them. The most important decision which has to be made is whether a perceptible correlation exists. In this context a correlation of 0.3 can be considered interesting and one of 0.5 to be of some importance. Correlation between two tests does not necessarily indicate a measure of causation. The comparison of means for both 1968-69 and 1969-70 indicates that students of higher than average verbal ability scored better on higher processes than on recall items whereas students of less than average verbal ability scored equally on both types. Students of higher than average verbal ability also scored significantly higher on both types of items than students of less than average verbal ability. It has long been thought that objective tests offer an opportunity to students of lower verbal ability to score better marks. The evidence of this present research indicates that, while this might very well be the case, it is clear that objective tests do not have a deleterious effect on the performance of the verbally more able students. The lack of significant difference in the performances of students from different home backgrounds was surprising. Early investigations showed that the children of routine workers tend, on the whole, to be less able verbally, and that objective testing might well suit them in examinations. The lack of significance in this investigation might well be explained by a "masking effect" of individuals in the middle range-say socio-economic groups III to VI. Examination of groups I and II compared with group VII might well show significant differences but was not possible in this case as the number of students was too small. It is worth considering that if, as suspected, conventional examinations place undue emphasis on the ability to recall plain facts, there is not much point in replacing them with objective tests if all that would then be measured was verbal ability or intelligence quotient.

AB - The relatively high correlation coefficients obtained between the Moray House Tests and the Co-operative Examinations are interesting, and it was this which prompted the investigation of performance on recall and higher processes items. The interpretation of correlation coefficients must, however, be approached with caution. Decisions on the nature of a good correlation must be made with other considerations, such as the nature of the sample, kept in mind. Correlations found in educational research tend to be rather smaller than those calculated in other areas of research. The reason for this is that, as the variables examined are usually complex in nature, it is unreasonable to expect high correlations among them. The most important decision which has to be made is whether a perceptible correlation exists. In this context a correlation of 0.3 can be considered interesting and one of 0.5 to be of some importance. Correlation between two tests does not necessarily indicate a measure of causation. The comparison of means for both 1968-69 and 1969-70 indicates that students of higher than average verbal ability scored better on higher processes than on recall items whereas students of less than average verbal ability scored equally on both types. Students of higher than average verbal ability also scored significantly higher on both types of items than students of less than average verbal ability. It has long been thought that objective tests offer an opportunity to students of lower verbal ability to score better marks. The evidence of this present research indicates that, while this might very well be the case, it is clear that objective tests do not have a deleterious effect on the performance of the verbally more able students. The lack of significant difference in the performances of students from different home backgrounds was surprising. Early investigations showed that the children of routine workers tend, on the whole, to be less able verbally, and that objective testing might well suit them in examinations. The lack of significance in this investigation might well be explained by a "masking effect" of individuals in the middle range-say socio-economic groups III to VI. Examination of groups I and II compared with group VII might well show significant differences but was not possible in this case as the number of students was too small. It is worth considering that if, as suspected, conventional examinations place undue emphasis on the ability to recall plain facts, there is not much point in replacing them with objective tests if all that would then be measured was verbal ability or intelligence quotient.

UR - http://www.scopus.com/inward/record.url?scp=33947085913&partnerID=8YFLogxK

M3 - Article

VL - 50

SP - 139

EP - 140

JO - Journal of Chemical Education

JF - Journal of Chemical Education

SN - 0021-9584

IS - 2

ER -