232

Research Notes

Br. J . educ. Psychol., 45,232-236, 1975

PERSONALITY, MODE OF ASSESSMENT AND STUDENT ACHIEVEMENT BY J. A. ROWELL AND V. J. RENNER (Department of Education, University of Adelaide, South Australia) SUMMARY. The EPI was administered to 136 full-time postgraduate Diploma in Education students. The relationshipsbetween personality, choice of method of assessment and achievement in four theory courses were investigated. Significant differences were found in each course favouring assessment totally by essay work (P< .05). In Theory of Education a significant interaction was found between neuroticism and assessment option, and in Educational Sociology extraverts achieved more highly than their relatively introverted peers. The hypotheses that introverts are more successful in structured courses and extraverts more so in relatively unstructured ones were tested and found to be supported.

INTRODUCTION The Eysenck Personality Inventory (EPI) has been widely used as an instrument for obtaining information from subjects concerning two major, purportedly basic, unrelated aspects of their personality : extraversion and neuroticism. In turn, the possibility that such information is of value to educators has been examined in numerous studies covering a wide range of ages, abilities, courses and institutions. For those interested in prediction the results to date have not been particularly encouraging, however, as is evident in several recent summaries. Entwistle (1972) for example, summarises the situation by saying “ It still requires considerable faith and imagination to see where these results may lead.” Kline and Gale (1971) and Elliott (1972) are no more helpful. Only k i t h (1973), who has utilised the technique of programmed instruction in an experimental attack on the possible relationship between persynality and achievement in what he refers to as “school-type learning situations, gives reason for optimism. k i t h maintains that introverts are more successful when they follow “ . . . a carefully sequenced, highly prompted structure of learning;” extraverts, he claims, appear to learn better in less formal ‘ discovery ’ situations. The question is still a long way from any generally accepted solution ; meanwhile there is another dimension of the problem which appears to have received much less, if any, research effort. This extra dimension is mode of assessment. In any system where options are available the question arises naturally of the possible relationships between personality, option chosen and achievement. The research reported here addresses this question. METHOD Sample. 136 postgraduate students, 72 females (median age 22 years, range 20-41 years) and 64 males (median age 223 years, range 20-40 years), representing 90.1 per cent of the 1973 enrolment of students to the one year full-time course leading to the Diploma in Education.

Test Instruments and Administration. All students and staff taking part in the investigation completed the Eysenck Personality Inventory, Form A, under the standard conditions described in the test manuaI (Eysenck and Eysenck, 1964). The majority of students for whom personality data were available completed the coursework assessments in psychology, sociology, history and theory of education. Assessment of coursework provided students with a choice of seven options, the choice in each subject being independent of that in any other subject. The options allowed assessment by essay work only, by essay work and/or an oral plus a formal written examination each providing marks in various weighted proportions, or by formal written examination only. Because of the small number of students electing to be assessed by some methods, however, it was necessary for the purposes of analysis to collapse the seven categories into three. These were (i) essay work only, (ii) assessment in which the formal written examination counted for at least 80 per cent of the total marks, (iii) other assessment options.

Research Notes

233

RESULTS AND DISCUSSION The mean extraversion (E), neuroticism (N) and lie (L) scores for the 123 students in the sample having L scores less than five are recorded in Table 1 by sex, separately, and as a group. The values are compared with the norms from the manual. TABLE 1 MEANSCORES AND STANDARD DEVIATIONS ON E.P.1. BY SEX.

I

Sample

Men students ...... Women students . . . . . .

56

67

I1

Neuroticism Mean(@

10.59 (5 *29) 12.49 (4.48)

I1 !: :f

Extraversion Mean(@

g::]

1

I

Lie Mean(@ 2.07 (1 *20) 2.00 (1.10)

.......... ...... In comparison with British undergraduates the Australian Diploma in Education students had higher N scores (P< .01, two-tailed) but there were no significant differences in E scores. Women students had, as expected, higher N scores than men (P< .05, two-tailed).

Personality and Achievement. Chi-square tests were used to examine the null hypotheses of no differences between personality groupings in students’ choice of method of assessment for each subject area considered separately. Because of the small number of students electing to be assessed by essay work only it was impossible to carry out such tests using the four EPI groups (NI, SI, NE, SE) and, instead, two tests were carried out for each course, separately categorising students as high and low N and high and low E, respectively-assignment to categories being determined by score relative to the mean of the group. None of the chi square values was significant. If choice of assessment is dependent on personality it was not apparent in this situation, given the limitations of measurement and analysis. Two-way analyses of variance (2 levels of N or E x 3 modes of assessment) were used to examine the null hypotheses of no difference in achievement of students differing in N o r E level, with the added dimension of comparative achievement between assessment options and the possibility of interactions. As mentioned above, numbers of students electing to be assessed by the essay method alone precluded use of the four EPI categories in these analyses. Separate analyses were undertaken for each course, rather than treating scores as replications by individuals, principally because of the probable differences between courses in terms of the nature of the methodology employed in their teaching and the possible interaction between personality and methodology (see k i t h , 1973 ; Entwistle, 1972).

Means and standard deviations of achievement scores, together with significant tests for the analyses of variance associated with each null hypothesis, are grouped in Table 2. Three interesting results emerge from Table 2. Firstly, it is apparent that, in general, students choosing to be assessed by the option involving essay work alone achieved higher marks than students choosing other ways of being assessed. The option allowing assessment by essay work alone was the most demanding of student time and the corresponding investment of student concern and effort is one possible explanation of the higher marks.

234

Research Notes

TABLE 2 MEANS,STANDARDDEVIATIONS AND SIGNIFICANCE TESTS ON 2 X 3 ANALYSISOF VARIANCE CCIMPARMQ PERSONALITY AND MODEOF ASSESSM.

-

Course

1 I 1 1

Hypotheses Tested

F

df

E Level . . . . . . . . . .

1

1,110

-1-1

1

Theory of Education

Low E

2

4 28 11

57.7 (5-0) 55 -9 (4 *3) 60 .O (5 ‘8)

18 24 11

55.3 (5-0) 57.4 (4-8) 60.5 (8.3)

1 1 2 2 2 I 32

61 *O (4.9) 59 *3(4 5 ) 62.4 (6.4)

1

HighE

History of Education

LowE

f

I

1 1 I 1 1

HighE

f

1

Educational Psychology

LowE

High E

Educational Sociology

LowE

I 1 1

2 3

,

,

i

HighE

1 ~

High N

Educational sociology

Interaction . . . . . . . . I *66 2,110

NS

-........

E Level

Assessment Option

1,109

NS

3.21

2,109

-05

I

2,109

NS

1

58-7 (6.3)

16 37 10

58.1 (6-2) 58.1 (4.7) 60-0 (8.5)

Assessment Option

16 26 14

53.9 (6.0) 55.7 (6.3) 61 *3(7 5 )

Interaction . . . . . . . .

37 11

54.4 (6-8) 56 a2 (7.2) 58.1 (7.4)

E Level

Assessment Option

56.7 7.0) 58 *6f6 *4) 63.6 (7.1)

Interaction . . . . . . . . N Level . . . . . . . . . .

1*45 1,110

NS

Assessment Option

-05

Interaction

......

. . . . . . . . 2.02

E Level

1,113

NS

4.25

2,113

-05

I a36

2,113

NS

- -. . . . . . . . . . 6.22

1,114

*05

4 -10

2.1 I4

.05

I

2,114

NS

-

I

21 2 3 1 2 11 2

57-0 ( 5 *7) 58 -0 (3 *6) 56-9 (7.1)

i3 Ij

4.35

2,110

56 -4 (4 -5) 55.5 (4.9) 63 -6 ( 5 *3)

Interaction . . . . . . . . 6.48

2,110

59.2 (6.2) 60.1 (4.3) 64.6 (6.7)

N Level . . . . . . . . . . 1 - 1 1

1,109

NS

Assessment Option

1

~

1

11

3 -57

2,109

.05

Interaction . . . . . . . . 1 *72 2,109

NS

1,113

NS

3 -29

2,113

*05

I

2,113

NS

3

11

1

15 27 16

55-9 (6.8) 58.0 (6.0) 61 -4 (7 *4)

N Level . . . . . . . . . .

17 36 8

56.1 (6.2) 56.5 (5.0) 59.5 (8.7)

Interaction . . . . . . . .

16 29 13

54.9 (8.5) 59-1 (6.4) 62.8 (6.6)

N Level . . . . . . . . .

2.22

1,114

NS

Assessment Option

4.21

2,114

-05

19 32

56.3 (5.4) 55.3 (7.0) 59.1 (8.5)

Interaction

.....

164

2,llr

NS

1

I

-01

-

60.8 (4.9) 58.0 (4.9) 61 *5 (7.2)

2 3

1

LowN

-0s

16 27 9

History of Education

Educational Psychology

2.1 10

3

HighN

LowN

4.36

I

I Theory of Education

I

Assessment Option

NS

2 3 1

2 3

-

-11

Assessment Option

I

-

Assessment options : 1 =80 per cent marks or more from formal written examination. 2=options other than 1 or 3. 3 =100 per cent marks from essay work.

-

235

Research Notes

The second result of interest is the significant interaction (P< .01) betweenNleve1 and assessment option which occurred only in theory of education. High N students achieved much higher scores than low N students in the 100 per cent essay option and, indeed, much higher scores than high or low N students in any other option. Following Welford’s (1965) ideas, a tentative theoretical explanation can be proposed. If N scores provide an indication of change in arousal level with stress, the opportunity to pace one’s efforts and be assessed on a number of essays rather than a ‘ one shot end of year examination m a y provide the high N student with one of the conditions necessary for him to realise his potential-a relatively non-stressful situation. However, this explanation is essentially a tentative post hoc rationalisation unsupported by data from the other subject areas. The third matter of significance is the fact that in educational sociology extraverted students achieved significantly higher scores (P< .05) than their more introverted peers. The result is contrary to a great deal of other data concerning the relationship between academic success and extraversion (Furneaux, 1962 ; Warburton, 1962 ; Rushton, 1966 ; Lavin, 1967 ; Leith, 1969 ; Leith and Davis, 1969), although Killingsworth (1972) in an attempted replication of the work of Furneaux (1962), and Leith and Wisdom (1970) have reported similar findings. k i t h (1973) has attempted to provide an explanation of his results in terms of an interactive relationship between method of instruction and the introversionlextraversiondimension of personality. According to Leith, “ . . . extraverts have a greater tolerance for ambiguity and lack of structure in the teaching situation, whereas introverts are more inclined to be responsive to unambiguous and clearly structured situations.” Unfortunately, kith’s (1973) paper was seen too late to question the majority of students about the degree of structuring in the courses in educational psychology and sociology. (The history and theory courses had a common core of teaching and content.) However, there was unanimity among staff, and those students who could be questioned, that the psychology course had a higher degree of structure than the sociology. On the basis of this, admittedly limited, information two hypotheses were proposed for testing : that introverts would achieve more highly in educational psychology than sociologyand that extravertswould achievemore highly in educational sociology than psychology. The results presented in Table 3 show both hypotheses to be supported. TABLE 3 ACH[EVeMENT SCORes OF

I

Personality

INTROVERTSAND EXTRAVERTS IN EDUCATIONAL PSYCHOLOGYAND socrouxtv.

,

Educational

i

Mean(@

!

Psychology

I

Educational

Sociology

I 1

I 1 1 p

Personality, mode of assessment and student achievement.

232 Research Notes Br. J . educ. Psychol., 45,232-236, 1975 PERSONALITY, MODE OF ASSESSMENT AND STUDENT ACHIEVEMENT BY J. A. ROWELL AND V. J. RENNE...
354KB Sizes 0 Downloads 0 Views