Make your own free website on Tripod.com

**************************************************************************************

Australian Psychologist, Vol. 14 No. 3, November, 1979, 337-344.

A Quick Measure of Achievement Motivation - Validated in Australia and Reliable in Britain and South Africa



J.J. Ray, University of New South Wales

ABSTRACT

A 14 item short form of the Ray Achievement Motivation scale was developed. When tested on seven samples from Sydney, London, Glasgow and Johannesburg it showed reliabilities of over .70 when applied to English speakers. It is also balanced against acquiescent response set and has validities well comparable with other longer scales. General population norms obtained in the four countries revealed the English, Scots and Australians to have similar levels of achievement motivation with South Africans significantly higher. The scale would appear to be unlike previous scales in that it was developed on general population rather than student samples.


INTRODUCTION

There seems to be considerable demand for a short self-report index of achievement motivation, suitable for use in survey research. Both Lynn (1969) and Smith (1973) have reported the development of such scales. The Lynn (1969) scale, however, was found in Ray (1971) and in O'Gorman (1974) to have reliabilities ("alpha") from .17 to .38, while the Smith scale showed a reliability ("alpha") of .56 in its development study and a reliability of .45 to .54 (split-half and test-retest) in a recent study by Opolot (1977). The standard of reliability accepted by Shaw-and Wright (1967) as suitable in a research instrument is .75. Clearly, a more reliable short scale is needed.

The present paper reports, then, the development of a short form of the Ray (1970, 1974, 1975) achievement motivation scale. This scale was originally developed from largely the same item pool as that used by Lynn (1969) and has been extensively validated on Australian samples (Ray, 1974; 1975). A slightly modified form of the scale (two items changed) was also validated by Beezhold (1975) in the Republic of South Africa. Beezhold divided the employees of a large South African insurance firm into those who had shown external signs of being achievement motivated (e.g. by enrolling in promotion-related study courses) and those who had not. Means for the two groups resulted in F ratios of 7.74 for whites and 4.00 for coloureds (p < .001).

STUDY I

This was an international study drawing on random cluster samples of the populations of London, Sydney and Glasgow. A total of 95 Sydney people, 100 Londoners and 100 Glaswegians were interviewed on their doorsteps. The male/female ratios in the three samples were respectively: 52/43, 53/47 and 52/48.

On the basis of the correlation (.64) shown in Ray (1974) between the original Ray (1970) scale and the Costello (1967) achievement motivation scale I, the longer scale was especially modified for the study by deleting six out of 30 items previously shown as correlating least with the scale total and replacing them with the first four negative items of the Costello scale. This gave a totally balanced 28 item pool, controlling for acquiescent response set, from which a short scale might be developed.

To enable cross-validation on the other samples, the item analyses were based solely on the London responses. A scale constructed on an English sample was also thought to have a greater potential for wide use because of the greater size of the English population.

The London responses were then subjected to the automatic item evaluation and deletion procedures of program ITRA (Ray, 1972). This program progressively deletes items having least correlation with all the other items and calculates the Cronbach (1951) coefficient "alpha" reliability statistic after each deletion. Alpha was found to be maximal at the 14 item length and this accordingly was the length finally chosen. Alpha at this length was .73.

When the Sydney and Glasgow data were re-scored for just the 14 items selected on the basis of the London results, alphas of .76 and .72 were observed. As alpha is the mean of all possible split-half reliabilities (Cronbach, 1951), it is in many ways an ideal reliability statistic and does hence give the 14 item scale a strong warrant of suitability for further use. For the reasons given at length in Nunnally (1967 p. 210 ff), test-retest correlations were not considered suitable as an index of reliability. See Tables 1 and 2 for the items and norms of the 14 item scale.

While 14 items is quite short for an adequate scale, it was felt that it might still be a little long for many doorstep applications. For this reason it was arbitrarily decided also to evaluate a ten item version of the scale. This was produced by deleting the four next-weakest items on the criterion of the item-to-total correlations. Items 3, 5, 7 and 13 of the 14 item scale were deleted. The alphas of this scale were .71 in London, .68 in Glasgow and .76 in Sydney. Table 2 gives its norms.

STUDY II

This was a validation study. It was carried out by rescoring the data from Ray (1974) for the shorter forms. The sample consisted of 75 Sydney males. The coefficients "alpha" observed were .77 (14-item) and .76 (10-item). The validity data available in this study was of four types: Peer-ratings, self-ratings, information on occupation and scores on other achievement motivation scales. Results from the first three of these are given in Table 3. The 14 item short form of the scale showed very similar validity properties to the longer forms. Most encouragingly, it gave the highest prediction of "actual achievement".

Although the validity correlations were in most cases not high, this is to be expected in validation exercises of this sort. Because human behaviour is in general multi-causal, no single variable can generally be expected to account for much of the variation in another. Furthermore, since motivation is an intrinsically very private matter, outside attempts to asses it tend to be very error-prone. An additional point concerns the validities of the criteria being used. If the criteria have limitations then the validity coefficients cannot be expected to be high. It may be noted that in the present case the average correlation between all the peer-ratings used was only .414 in spite of their very high level of conceptual similarity. The important thing is to show that two sources of information (tests and criteria) give convergent results at highly significant levels. This was done on the present occasion.


Table 1

The items of the short form of the Ray-Lynn AO scale. Response options are "Yes", (scored 3), "?" (scored 2), "No" (scored 1). Items marked "R" are to be reverse-scored (e.g. "1" becomes "3") before addition to get the overall score.

1. Is being comfortable more important to you than getting ahead? R
2. Are you satisfied to be no better than most other people at your job? R
3. Do you like to make improvements to the way the organization you belong to functions?
4. Do you take trouble to cultivate people who may be useful to you in your career?
5. Do you get restless and annoyed when you feel you are wasting time?
6. Have you always worked hard in order to be among the best in your own line? (school, organization, profession).
7. Would you prefer to work with a congenial but incompetent partner rather than with a difficult but highly competent one? R
8. Do you tend to plan ahead for your job or career?
9. Is "getting on in life" important to you?
10. Are you an ambitious person?
11. Are you inclined to read of the successes of other rather than do the work of making yourself a success? R
12. Would you describe yourself as being lazy? R
13. Will days often go by without your having done a thing? R
14. Are you inclined to take life as it comes without much planning? R


Table 2.

Population norms for three forms of a self-report achievement motivation inventory.

...............................................Glasgow...............London...............Sydney

28 item mean and S.D..............62.94 (7.93)...........63.51 (7.46).......62.07 (9.13)
14 item mean and S.D..............31.33 (5.64)...........32.45 (5.71).......31.44 (5.83)
10 item mean and S.D..............21.12 (4.57)...........22.54 (4.73)....... 21.90 (4.84)

Subdivided by sex:

Males (28 item form)................63.96 (8.44)...........64.20 (7.46).......62.63 (8.82)
Males (14 item form)................32.61 (5.66)...........33.49 (5.51).......32.17 (5.32)
Females (28 item form)............61.83 (7.17)...........62.72 (7.38).......61.39 (9.45)
Females (14 item form)............29.93 (5.29)...........31.27 (5.69).......30.55 (6.29)
Males (10 item form)................22.23 (4.41)...........23.50 (4.67).......22.33 (4.61)
Females (10 item form)............19.91 (4.44)...........21.44 (4.55).......21.40 (5.06)

Sample total n.............................100.........................100....................95


Table 3.

Validity correlations observed on 75 Sydney males for four forms of a self-report inventory of achievement motivation.

.......................................................30 items....28 items....14 items...10 items

Peer ratings:

Task orientation................................. .25............ .30............ .28.......... .21
Success orientation............................ .33............ .37........... .39.......... .44
Need for achievement........................ .32............ .36............ .30.......... .33
Achievement orientation.................... .32............ .37............ .33.......... .34
Actual achievement........................... .17............ .26............ .32.......... .32


Self rated achievement motivation..... .58............ .58............ .53.......... .53
Occupation (scored: 1 = Manual; 2 = non-Man.)
........................................................... .31........... .32............ .34........... .40
Reliability ("alpha")............................ .74........... .81........... .77........... .76
Mean..................................................67.40.........64.05.........32.32........22.36
S. D.....................................................8.60...........9.42...........5.81..........4.80

Using a contrast-wise error-rate approach, all correlations above .190 are significant at the .05 level (one-tailed).

The data from the Ray (1974) study also contained scores on a variety of other self-report achievement motivation indices plus data from one projective test - the French (1958) Test of Insight. This data was used to provide concurrent validation for the new scales. The other tests were a forced-choice test -- the Mukherjee (1965) SCT (Sentence Completion Test) - the McReynolds and Guevara (1967) Success-seeking and Avoidance of Failure tests, the Costello (1967) Task-orientation and Success-orientation tests and the Mehrabian (1968) test of resultant achievement motivation. A short social desirability scale from Greenwald and Satow (1970) as also included as a check on set artifacts.

Table 4 gives the relevant correlations. It will be seen that they are generally satisfactory. As is usual in self-report tests, there was little correlation with the projective test; however, the projective test also failed to predict any of the peer rating validity criteria. The lack of correlation between the self-report and projective tests would, in other words, appear to be explained by the relatively poor validity of the projective test. Table 4 also gives correlations with two of the most central of the peer ratings for all tests included in the battery to enable assessment of comparative validity.


Table 4.

Correlations between four forms of the Ray Achievement motivation scale and other scales.

.....................................................................................................Peer Ratings
...............................30 items...28 items...14 items...10 items ....Task O....Succ O

French Test of
Insight (projective).... .02.......... .03........... .04............ .07............ .09....... .15
Mukherjee S.C.T........ .52.......... .50........... .51............ .52............ .19....... .30
Costello II (Success
orientation)............... .10.......... .04.......... -.03........... -.00........... -.07......-.05
Costello I (Task
orientation)............... .64.......... .75........... .78............ .72............. .30...... .29
McReynolds &
Guev. Failure
Avoidance................. .23.......... .25........... .32............ .32............. .14...... .21
McReynolds &
Guev. Success
Seeking..................... .43.......... .45........... .53........... .55............. -.01...... .37
Mehrabian scale........ .38.......... .41........... .51........... .54............. .20....... .35
Social Desirability
Response Set............ .17.......... .19........... .13........... .12.............. .01...... .04


STUDY III

The remaining issue appeared to be the effect of context on the new short tests. Would they be as reliable without being surrounded by the items of the longer scale? Studies III and IV were carried out to check on this.

In this study, the 14 item form was administered to a random cluster sample of 145 Sydney people (78 females and 66 males). This was again a doorstep sample selected by methods modelled on those used by the commercial public opinion polls.

The reliability, in fact, rose slightly -- to .79. The mean score and its standard deviation at 32.17 and 6.23 were also very similar to the values previously observed. Clearly, then, the short scale functions very similarly even outside its original context.

STUDY IV

In this study the effect of context on the 10 item form was studied. The sample was gathered by having students administer the scale to people they knew under the constraint that non-students were to be sampled and people in the humbler occupations were to be preferred. As hoped, the effect of these constraint was to produce a sample which did not in fact differ significantly in demographic characteristics (age, sex, occupation and education) from the sample of Study III. With an N of 87 (40 males and 47 females) this was, then, an adequate quota sample of the Sydney area.

The reliability ("alpha") observed for the scale was .73 with a mean and S. D. of 22.41 and 4.60. Although the reliability has in this case dropped rather than risen, the degree of difference is still not great. Again it has been demonstrated that a short form functions quite well without the context of the other items amid which it was originally set.

STUDY V

Given that the short scale had at this stage been fairly well tested, it seemed useful to try it out in some substantive application. The application chosen was to study levels of achievement motivation in what is arguably one of the most "different" English-speaking cultures today -- that of the Republic of South Africa. What is the motivational effect of being a member of a white elite among a majority of blacks?

Theoretically, one could predict that being born into an elite would reduce achievement motivation. Why strive when one is born on top? On the other hand one could argue that consciousness of one's elite role would make one strive all the harder to live up to it and to justify the accident of one's birth. Only testing could possibly decide which effect predominates.

The 14 item scale was then included in a short questionnaire given to 100 South Africans (52 males and 48 females). The sample was a random cluster sample taken in the Johannesburg Greater Metropolitan Area using sampling methods identical to those used in studies I and III. As in the British samples of Study I, all interviewing was carried out by the author personally. As the British sampling took place in October 1977 and the Johannesburg interviewing took place in December 1978, however, contemporaneity of interviewing was only approximate.

Although the South African sample did have predictably fewer respondents in blue-collar occupations, the differences on the four demographic characteristics of age, sex, occupation and education between the present sample and that of Study III were not in fact significant. Whites only were sampled.

The coefficient "alpha" reliability observed was .67 with a mean of 35.22 and an S.D. of 4.75. As this was low, the sample was divided into those (n = 57) whose native language was English and those for whom English was a second language. The alpha for the English was .72 (mean of 35.14 and S.D. of 5.02) while for the Afrikaners it was .57 (mean of 35.64 and S.D. of 4.14). (There were also five respondents whose native language was neither English nor Afrikaans. They were, of course, too few to support a separate analysis.) All in all, the reliability among the English-speakers was virtually identical to that observed in the U.K.

The t for the difference between the Johannesburg mean and the Sydney mean of Study III was 4.06 (p <.01). Thus in spite of the considerable similarities between South Africa and Australia in terms of climate, heredity and culture, it would seem that the South African social situation does lead to higher achievement motivation.

The reduced reliability of the 14 item scale for the Afrikaners is to be expected of a scale administered in a language not fully familiar to the respondent. Because the items are less well comprehended, they are less consistently responded to. Clearly the scale should be translated if it is to be administered to such populations.

DISCUSSION

In testing on two British and three Australian samples, the 14 item achievement motivation scale has been shown to have uniformly satisfactory reliability. It has also been shown to have validity as shown by correlations with occupation, peer-ratings and self-ratings. An even shorter 10 item form has very similar characteristics.

Some points of interest that emerge from comparing the Ray scales with other scales are their generally greater validity in terms of peer ratings (compare the last two columns of Table 4 with the first two rows of Table 3) and the wide range of other scales which correlate with them (See Table 4). The two peer ratings given in Table 4 correspond to what seems a fairly basic division of achievement motivation into intrinsic versus extrinsic or task versus success orientation (see Featherman, 1972). This may in some senses correspond to McClelland's distinction between achievement and power motivation.

One not previously documented finding to come out of Study II above was the generally high validity of the Costello scale I. With an "alpha" of .75, this 10 item balanced scale was revealed as being most valid of all against the criterion of peer rated task orientation, though well down on peer rated success orientation. Because of its U.S. origin, it would hence be a useful resource in studies involving comparisons between the U.S.A. and Australia.

The attraction of the new 14 item scale would then lie in:

1) Its brevity - with doorstep administration time from 5 to 10 minutes;
2) Its fairly consistent reliability internationally. This would be particularly useful in cross-cultural research involving other English-speaking cultures;
3) The fact that few if any achievement motivation scales appear heretofore to have been validated for use with Australian general population samples. This latter feature would give it particularly strong claims to use in future Australian research;
4) Its generally superior validity.

For higher reliabilities, other scales of overseas origin such as the 20 item scale from the Jackson (1967) PRF could perhaps be turned to but it must be noted that this scale is copyrighted for administration as part of a long personality inventory and has not been designed for doorstep use. Like many scales derived from large American personality inventories, it appears, in fact, to have been constructed and normed entirely on a U.S. college student population. Its applicability to even a U.S. random population sample cannot therefore be guaranteed.

The new scale has already enabled cross-cultural findings of some interest. Despite many reasons one could think of to the contrary, Table 2 shows that Australia, England and Scotland do not in fact differ on average levels of achievement motivation. The higher levels in South Africa are, then, doubly in need of explanation. This may suggest that greater degrees of social inequality lead to higher levels of personal achievement motivation.

REFERENCES

Beezhold, M.A. A study of job orientation & motivation in different groups of white & coloured employees. Unpublished Ph.D. thesis, University of Stellenbosch, 1975.

Costello, C.A. Two scales to measure achievement motivation. l. Psychology, 1967, 66, 231-235.

Cronbach, L.J. Coefficient alpha and the internal structure of tests. Psychometrika, 1951, 16, 297-334.

Featherman, D.L. Achievement orientations and socioeconomic career attainments. American Sociological Review, 1972, 37, 131-143.

French, E. Development of a measure of complex motivation. In J.W. Atkinson (Ed.) Motives in fantasy, action and society Princeton, N.J.: Van Nostrand, 1958.

Greenwald, H.J. & Satow, Y. A short social desirability scale. Psychological Reports, 1970, 27, 131-135.

Jackson, D.N. Personality research form manual N.Y.: Research Psychologists Press, 1967.

Lynn, R. An achievement motivation questionnaire. British Journal of Psychology, 1969, 60, S29-S34.

McReynolds, P. & Guevara, C. Attitudes of schizophrenics and normals towards success and failure. Journal of Abnormal and Social Psychology, 1967, 72, 303-310.

Mehrabian, A. Male and female scales of the tendency to achieve. Educational & Psychological Measurement, 1968, 23, 493-502.

Mukherjee, B.N. A forced choice test of achievement motivation. Journal of the Indian Academy of Applied Psychology, 1965, 2, 85-92.

Nunnally, J.C. Psychometric theory N.Y.: McGraw Hill, 1967.

O'Gorman, J.G. On the validity of Lynn's achievement motivation questionnaire. British Journal of Social and Clinical Psychology, 1974, 13, 209-210.

Opolot, J. A. Reliability and validity of Smith's quick measure of achievement motivation scale. British Journal of Social and Clinical Psychology, 1977, 16, 395-396.

Ray, J.J. (1970) Christianism.... The Protestant ethic among unbelievers. J. Christian Education, 13, 169-176.

Ray, J.J. (1971) Correspondence: Regarding the Lynn n-Ach test. Bulletin British Psychological Society, 24, 352.

Ray, J.J. (1972) A new reliability maximization procedure for Likert scales. Australian Psychologist 7, 40-46.

Ray, J.J. (1974) Are trait self-ratings as valid as multi-item scales? A study of achievement motivation. Australian Psychologist 9, 44-49.

Ray, J.J. (1975) A behavior inventory to measure achievement motivation. J. Social Psychology 95, 135-136.

Shaw, M.E. & Wright, J.M. Scales for the measurement of attitudes N.Y.: McGraw Hill, 1967.

Smith, J.M. A quick measure of achievement motivation. British Journal of Social and Clinical Psychology, 1973, 12, 137-143.



POST-PUBLICATION ADDENDUM

Replication is one of the cornerstones of science. A new research result will normally require replication by later researchers before the truth and accuracy of the observation concerned is generally accepted. If a result is to be replicated, however, careful specification of the original research procedure is important.

In questionnaire research it has been my observation that the results are fairly robust as to questionnaire format. It is the content of the question that matters rather than how the question is presented (But see here and here). It is nonetheless obviously desirable for an attempted replication to follow the original procedure as closely as possible so I have given here samples of how I presented my questionnaires in most of the research I did. On all occasions, respondents were asked to circle a number to indicate their response.




Go to Index page for this site

Go to John Ray's "Tongue Tied" blog (Backup here or here)
Go to John Ray's "Dissecting Leftism" blog (Backup here or here)
Go to John Ray's "Australian Politics" blog (Backup here or here)
Go to John Ray's "Gun Watch" blog (Backup here or here)
Go to John Ray's "Education Watch" blog (Backup here or here)
Go to John Ray's "Socialized Medicine" blog (Backup here or here)
Go to John Ray's "Political Correctness Watch" blog (Backup here or here)
Go to John Ray's "Greenie Watch" blog (Backup here or here)
Go to John Ray's "Food & Health Skeptic" blog (Backup here)
Go to John Ray's "Leftists as Elitists" blog (Not now regularly updated -- Backup here)
Go to John Ray's "Marx & Engels in their own words" blog (Not now regularly updated -- Backup here)
Go to John Ray's "A scripture blog" (Not now regularly updated -- Backup here)
Go to John Ray's recipe blog (Not now regularly updated -- Backup here or here)

Go to John Ray's Main academic menu
Go to Menu of recent writings
Go to John Ray's basic home page
Go to John Ray's pictorial Home Page (Backup here)
Go to Selected pictures from John Ray's blogs (Backup here)
Go to Another picture page (Best with broadband)