This is the author’s version of a work that was submitted/accepted for pub-

lication in the following source:

McKenzie, Kirsten & Schweitzer, Robert D.

(2001)
Who succeeds at university? factors predicting academic performance in
first year Australian university students.
Higher Education Research & Development, 20, pp. 21-33.

This file was downloaded from: https://eprints.qut.edu.au/56040/

c Copyright 2001 Routledge

Notice: Changes introduced as a result of publishing processes such as

copy-editing and formatting may not be reflected in this document. For a
definitive version of this work, please refer to the published source:

https://doi.org/10.1080/07924360120043621
 Factors predicting academic performance 1

Running head: FACTORS PREDICTING ACADEMIC PERFORMANCE

Who Succeeds at University?

Factors Predicting Academic Performance in

First Year Australian University Students

Kirsten Vallmuur and Robert Schweitzer

Queensland University of Technology, Brisbane

Address for correspondence:

Kirsten Vallmuur

School of Psychology, L Block

Queensland University of Technology

Carseldine Campus, Carseldine , QLD, 4034

 Factors predicting academic performance 2

Abstract

With the increasing diversity of students attending university, there is a growing interest in

the factors predicting academic performance. This study is a prospective investigation of

the academic, psychosocial, cognitive, and demographic predictors of academic

performance of first year Australian University students. Questionnaires were distributed

to 197 first year students 4 to 8 weeks prior to the end of semester exams and overall grade

point averages were collected at semester completion. Previous academic performance was

identified as the most significant predictor of university performance. Integration into

university, self efficacy, and employment responsibilities were also predictive of university

grades. Identifying the factors that influence academic performance can improve the

targeting of interventions and support services for students at risk of academic problems.
 Factors predicting academic performance 3

Over the last decade, higher education in Australia has seen a shift from elite to

mass education. With the reforms of the late 1980's, equity and access for all have been a

primary focus of Australian universities. In a review of higher education, West (1998)

stated "central to the vision is ensuring that no Australian is denied access to a high quality

of education at any level merely because of his or her social background or financial

circumstances" (p. 2). The Organisation for Economic Cooperation and Development

(1997) reported that between the years from 1983 to 1995, university enrolments in

Australia rose by 70 per cent.

Accompanying this growth in higher education is an increasing diversity amongst

the student population. Students from different social and cultural backgrounds, with

different experiences and varying levels of education bring with them different needs and

academic potential. The challenge for Australian universities is to recognise this diversity

of needs and cater for this changing and heterogeneous population of students. Power,

Robertson, and Baker (1987) stated "the stress should not only be on admitting a wider

range of students, but also on giving them the support and help needed to ensure a

reasonable chance of success" (p. 3).

The aim of this study was to undertake a prospective investigation of the academic,

psychosocial, cognitive, and demographic predictors of academic success in an Australian

university context. A broad range of factors were examined in recognition of the diversity

of student needs. This study examined the major predictors of academic success identified

in previous research, with a view to developing a model which could be used to identify

students at risk of academic problems.

Prospective research of this kind is crucial during the current changes that

Australian higher education is undergoing. This research has implications for both

university policy and student support services. Through identifying the factors that
 Factors predicting academic performance 4

influence the academic performance of students, universities are better equipped to target

interventions and support services to meet the needs of at risk students, with a view to

optimising the retention of students and maintaining standards.

The sections following will provide a review of previous research on academic,

psychosocial, cognitive and demographic predictors of academic performance.

Academic Predictors of Academic Performance

The two major academic predictors of performance at university identified in the

literature are previous academic performance and study skills.

An extensive amount of research has shown support for the relationship between

previous academic performance and university performance (Baker and Siryk, 1984; Clark

and Ramsay, 1990; Everett and Robins, 1991; Gerdes and Mallinckrodt, 1994; Pascoe,

McClelland, and McGaw, 1997; Power, Robertson, and Baker, 1987). Power, Robertson,

and Baker (1987) reported that the correlation between secondary school grades and Grade

Point Average (GPA) at university is generally about 0.5.

However, the predictive capacity of secondary school grades is different for

different individuals and groups. Power, Robertson, and Baker (1987) found that

secondary school grades are not as good predictors for mature age student's performance as

they are for school leaver's performance, and female students with the same secondary

school grades as male students consistently outperform their male counterparts (p. 12 –

13). The method of entry into university and the ease with which entry could be made into

university has also been found to affect the predictive capacity of secondary school grades

(Pascoe, McClelland, and McGaw, 1997). Previous academic performance does not

sufficiently account for the variability in student's university grades. Other academic

factors such as study skills have been found to affect university grades.

Study skills have been found to influence academic performance. Pantages and
 Factors predicting academic performance 5

Creedon (1975) found that students with poor study habits are more likely to withdraw

from university or to have academic adjustment problems in the transition from high

school to university. Abbott-Chapman, Hughes, and Wyld (1992) found that “those

students classified as “academics”, for instance, whose profiles include an active “liking

for study”, are more likely to perform well at university than those students who don’t have

this motivation” (p. 17).

Fortunately, study skills are not absolute and unchangeable. Study skills can be

improved through training and courses and therefore lead to improved GPA's. Cone and

Owens (1991) found that students who participated in a study skills course obtained higher

GPA's than predicted by previous academic performance. Cone and Owens (1991) stated

“college freshman who are deficient in basic study skills are likely to become highly

anxious in their new academic setting thereby further impairing their potential for

academic success" (p. 1211). This study must be interpreted with caution, as the

researchers neglected to measure study skills prior to or following the study skills course.

Improvement in student’s GPA’s may be due to other factors associated with participation

in the course.

Two academic factors have been highlighted as predictive of university success:

secondary school grades and study skills. These variables have been found to contribute

moderately to this prediction. Secondary school grades explain the largest portion

(approximately 0.5) of the variance in GPA, with several studies showing study skills

explain a small amount of unique variance over and above this. Thus, other factors must

be important in the prediction of university success if academic factors account for no

more than half of the variance in GPA.

Psychosocial Predictors of Academic Performance

There is a vast body of research that highlights the importance of psychosocial

 Factors predicting academic performance 6

variables in predicting academic performance (Gerdes and Mallinckrodt, 1994; Lecompte,

Kaufman, and Rousseeuw, 1983; Lecompte, Kaufman, Rousseeuw, and Tannin, 1983;

Rickinson and Rutherford, 1995; Rickinson and Rutherford, 1996; Terenzini and

Pascarella, 1978; Tinto, 1975; Wince and Borden, 1995). Several psychosocial predictors

of academic performance have been identified: student institution integration, commitment

to university, satisfaction with university, financial difficulty, career orientation, social

support, and psychological health.

A major contributor to the study of the relationship between psychosocial factors

and academic performance is Vincent Tinto. Tinto (1975) developed a student integration

model which emphasises the academic and social integration, and the educational and

institutional commitment of the student as the most important predictors of student

attrition. Academic and social integration are defined as “a subjective sense of belonging

and fitting in....a sense of compatibility or dissonance with the University and its students”

(Spady, 1971, p. 44). The model suggests that a match between the academic ability and

motivation of the student with the social and academic qualities of the institution foster

academic and social integration into the university system. This leads to the development

of two commitments: an educational commitment and an institutional commitment.

Commitment refers to both the degree of importance the student places on achieving their

academic and career goals, and the degree to which the student identifies and is committed

to the university at which they are studying. According to the model, if the student is not

integrated into the university, they will develop a low commitment to university. While

this model does not deny the academic history of the student, it does not see academic

history as predictive of student withdrawal. Instead, academic history influences the

process of student integration. It is the degree of integration and commitment experienced

by the student that influences that student’s decision to withdraw or continue their
 Factors predicting academic performance 7

university education.

Tinto’s model has gained considerable support in the literature. Terenzini and

Pascarella (1978) found the most significant predictors of student attrition were academic

and social integration variables, with previous academic performance and personality

variables accounting for only four percent of the variance in attrition status. Power,

Robertson, and Baker (1987) established “the most important variable distinguishing

between those who proceed successfully to second year and those who dropout or fail, is

course commitment”(p. 40). Lecompte, Kaufman, and Rousseeuw (1983) discovered the

less students made use of student services (academic/social integration), the more likely a

student was to withdraw from university.

However, the prediction that social integration is associated with academic

achievement is arguable. Research has shown that social integration into the university is

not necessarily beneficial for achieving high grades. McInnis, James and McNaught

(1995) found a higher percentage of students achieving average marks worked in social

groups to study, while students achieving the highest and lowest marks were less social in

their academic work. Tinto (1975) also recognised that social integration is not always

beneficial for academic achievement:

Insufficient social interaction seems to lead primarily to voluntary withdrawal,

whereas excessive social interaction may, in some cases, lead to dropout if the

group with whom one associates is itself disinclined toward academic achievement.

(p. 109)

Unfortunately, Tinto’s model does not take this recognition of the harms of social

integration into account, offering a more generalised explanation of student attrition. This

general view does not fit for all students. McInnis, James and McNaught (1995) explain

“In some student cultures, social integration is contingent on not being diligent – being a
 Factors predicting academic performance 8

‘conch’ (conscientious student), or at least, a certain type of ‘conch’, can be a social

liability”(p. 39).

Research has also shown that other psychosocial factors influence academic

performance. These include satisfaction with university, financial situation, career

orientation, and social support. Rickinson and Rutherford (1996) found dissatisfaction

with the course of study was the reason most commonly endorsed for leaving university.

Wince and Borden (1995) found satisfaction with university was related to both higher

GPA's and lower withdrawal rates.

Lecompte, Kaufman, and Rousseeuw (1983) found financial difficulties was also a

common reason for leaving university. However, Lecompte, Kaufman, Rousseeuw, and

Tassin (1983) found financial difficulty was not significantly related to academic

performance. Other studies have also found a negative correlation between financial

difficulties and student retention (Braxton, Brier, and Hossler, 1988; Pantages and

Creedon, 1975), but few studies have examined the relationship between financial

difficulties and academic performance.

In terms of career orientation, Himelstein (1992) reported that students with a clear

career orientation achieved higher GPA's and were less likely to withdraw from university

than students lacking a clear career orientation. Williams (In McInnes, James and

McNaught, 1995) found students with clear objectives for university were more likely to

achieve higher grades than those students with undefined objectives (p. 28).

Social support has also been found to influence academic performance. Several

researchers have found that the presence of a person who provides strong support and

support from family or spouse are important predictors of student retention and academic

success (Gerdes and Mallinckrodt, 1994; Pantages and Creedon, 1975; Tracey and

Sedlacek, 1982).
 Factors predicting academic performance 9

Psychological health variables have gained a limited amount of attention in the

academic performance literature. Few studies have included measures of psychological

health as variables in academic performance research, and even fewer have looked solely

at the relationship between psychological health and academic performance. The findings

from these studies support the relationship between psychological health and academic

performance.

Houston (1971) and Lecompte, Kaufman, Rousseeuw, and Tassin (1983), in a

prospective study, found that student’s reporting high anxiety at the start of the academic

year had significantly poorer grades at the end of the academic year than their less anxious

peers. Lecompte, Kaufman and Rousseeuw (1983) found that student’s experiencing more

frequent episodes of depression had a higher dropout rate than students experiencing fewer

depressive episodes. Szulecka, Springett, and de Pauw (1987), also utilising a prospective

methodology, found that higher levels of psychological disturbance, such as depression,

anxiety and somatic complaints, were related to a significantly higher incidence of

withdrawal from university.

Cognitive Appraisal as a Predictor of Academic Performance

Cognitive appraisal research tends to fall into one of two categories in academic

performance literature: studies of self efficacy and studies of attributional style. Self-

efficacy needs to be distinguished from attributional style.

Self-efficacy refers to “the belief’s about one’s ability to perform successfully a

given task or behaviour” (Lent, Brown and Larkin, 1987, p. 293). In academic

performance literature, self efficacy has referred to a belief that one will achieve good

grades in a given course or subject (Gerdes and Mallinckrodt, 1994; Lecompte, Kaufman,

and Rousseeuw, 1983; Lecompte, Kaufman, Rousseeuw, and Tassin, 1983; Lent, Brown

and Larkin, 1984, 1987).

 Factors predicting academic performance 10

Attributional style refers to a general appraisal of events and a belief about the

causes of events. From an attributional style perspective, individuals explain events in

terms of internality, stability, and pervasiveness, and the way individuals explain events

can be used to classify people as optimists or pessimists (Peterson and Barrett, 1987;

Pierce and Henry, 1993). In academic performance literature, attributional style has been

examined in relation to the characteristics of high and low achievers, that is, are they

optimists or pessimists in their orientation to the world in general.

Thus, attributional style is a general appraisal of events in terms of the dimensions

of pessimism and optimism, whereas self-efficacy is a belief about one’s ability in terms of

a specific event such as achieving high grades in a course. While it is recognised that there

is some similarity between the two aspects of cognitive appraisal, the distinction between

self-efficacy as specific appraisal and attributional style as general appraisal will be used in

this research.

Self-efficacy has been found to be predictive of university grades. A belief that one

will perform successfully in a given course predicts actual successful performance in that

course. Lent, Brown and Larkin (1987) found “self-efficacy added significant unique

variance beyond measures of objective ability and achievement in predicting subsequent

academic performance and persistence” (p. 293). Lecompte, Kaufman, Rousseeuw, and

Tassin (1983) and Lecompte, Kaufman, and Rousseeuw (1983) found that an expectation

of academic success (self-efficacy) has a highly significant positive relationship with

actual academic success and with low withdrawal rates. Gerdes and Mallinckrodt (1994)

reported students with high grades were more confident about their own ability to achieve

academic success than students with poor grades and withdrawers from university.

One of the more recent trends in academic performance research has focused on

attributional style as a predictor of academic achievement. Attributional style has its roots
 Factors predicting academic performance 11

in the theory of ‘learned helplessness’ (Abramson, Seligman, and Teasdale, 1978).

“Learned helplessness is the giving up reaction, the quitting response that follows from the

belief that whatever you do does not matter” (Tominey, 1996, p. 5).

Pierce and Henry (1993) explain attributional theory in terms of pessimism and

optimism:

According to this reformulated learned helplessness theory, those who attribute

failure to internal-stable-global causes have a pessimistic attributional style and are

more likely to display symptoms associated with learned helplessness such as not

trying when faced with failure. Those who attribute failure to external-unstable-

specific causes have an optimistic attributional style and are expected to continue to

work in the face of failure. (p. 5)

Little research has been undertaken in the area of attributional style and academic

success within a general university student populations. A study by Peterson and Barrett

(1987) utilising a small sample of 87 first year students examined attributional style and

academic performance in a university population. They found a significant negative

relationship between pessimistic attributional style and first year GPA.

Attributional style has been found to significantly predict grades over and above

that predicted by traditional academic measures. Seligman (1991) measured student’s

attributional styles upon entering university and found that optimistic students achieved

significantly better grades than predicted by the traditional measures (high school grades

and achievement tests) and pessimistic students did more poorly than predicted by

traditional measures.

A caveat to some of these studies is the use of an academic modification of the

ASQ. This measure of attributional style was developed and used by Peterson and Barrett

(1987) without evidence of testing the psychometric properties of the measure. Since then,
 Factors predicting academic performance 12

other studies have also used the academic version of the ASQ without any reported

evidence of the measure having sound psychometric properties (reliability or validity).

The academic modification of the ASQ used in some studies cites only negative

events, unlike the original ASQ that cites both negative and positive events. The negative

slant of the academic version of the ASQ may influence the results in a more pessimistic

direction than would be expected from the ASQ. Many studies have reported that the ASQ

dimension of CoNeg (total of all negative events) produces more significant results than

the often insignificant ASQ dimension of CoPos (total of all positive events) (Pierce and

Henry, 1993; Seligman and Schulman, 1986; Tominey, 1996). Thus, the academic version

of the ASQ in measuring only the negative side of attributional style, may unduly influence

the significance of the results.

Demographic Predictors of Academic Performance

Demographic variables that have been found to be important influences on

academic performance are age, gender, employment responsibilities, and student workload.

Studies have reported inconsistent results when measuring the relationship between age

and academic achievement. Some studies show a significant negative relationship between

age and academic achievement, in that school leavers achieve higher results and are more

likely to persist at university than mature age students (Clark and Ramsay, 1990; Pantages

and Creedon, 1975). Other studies have found that mature students, having a clearer career

orientation and lower integration needs, are more likely to achieve higher academic results

(McInnis, James and McNaught, 1995).

Employment responsibilities have been found to influence student retention.

Pantages and Creedon (1975) found full-time students who worked more than 15 hours per

week were more likely to withdraw than full-time students who worked less than 15 hours

per week. McInnis, James and McNaught (1995) state: “The pressures of part-time, and in
 Factors predicting academic performance 13

a minority of cases, full-time work, make it extremely difficult for some students to fulfil

course expectations” (p. 69). There has been little research into the effect of employment

on academic performance, but it could be argued that full-time students with greater

employment responsibilities would not perform as well as full-time students with little or

no employment responsibilities.

There has been little research examining the influence of student workload on

academic performance. While research has identified that part-time students are often less

integrated into university (McInnis, James and McNaught, 1995), research on part-time

students has not been expanded to examine academic performance.

Rationale for Study

Little prospective research has been undertaken on the predictors of academic

performance which integrates both academic and non-academic predictors. The majority

of studies have employed a cross-sectional or retrospective methodology. Research tends

to focus on the relationship between academic performance and one of the three broad

areas of academic factors, psychosocial factors, or cognitive appraisal. The few studies

that have examined more than one of these areas have been directed at the study of student

attrition rather than academic performance (Gerdes and Mallinckrodt, 1994; Pascarella,

Duby, Miller, and Rasher, 1981; Spady, 1971; Terenzini and Pascarella, 1978; Tinto,

1975). The majority of these studies have been based on Tinto’s Student Integration

Model, which has been criticised and contradicted by some researchers (McInnis, James

and McNaught, 1995; McKeown, Macdonell, and Bowman, 1993). The influence of

demographic variables is often excluded, or not examined adequately in many studies.

There has been no systematic research which examines both academic and non-academic

predictors of academic performance and the relationship between these predictors.

Moreover, there has been little published research of either academic or non-academic
 Factors predicting academic performance 14

predictors, let alone an integration of these, performed within an Australian tertiary

context.

This prospective study will examine the relationship between academic,

psychosocial, cognitive, and demographic variables, and the academic performance of first

year Australian university students. The variables are based on factors identified in

previous research as being important predictors of academic performance. Academic

performance is based on the first semester Grade Point Average (GPA) of students. The

aim of the study is to identify the variables within each of the four factors that affect

academic performance, with a view to develop a model which could be used to identify

students at risk of academic problems. Based on previous research, several hypotheses are

proposed.

1. Higher grades in secondary school and well developed study skills will be associated

with higher university grades

2. Integration, commitment, satisfaction with university, social support, lack of financial

difficulty, and a clear career orientation will be related to higher university grades.

3. Psychological health will be related to higher university grades.

4. High self efficacy and optimism will be related to higher university grades.

5. Full-time students with limited employment responsibilities will have higher GPA’s

than full-time students with more demanding employment responsibilities.

Method

Participants

A sample of 197 first year university students from the Faculties of Science (n =

149) and Information Technology (n = 48) in a large urban commuter-based university

volunteered to participate in the study. The sample included 103 males and 94 females
 Factors predicting academic performance 15

with a mean age of 21.24 years.

Materials

A questionnaire was developed to measure academic variables, psychosocial

variables, cognitive appraisal, and some demographic variables. The academic variable

that was measured in the questionnaire was self reported study skills. Self reported study

skills were assessed using the Academic Problems sub-scale from the College Adjustment

Scales Inventory. This scale measures the extent of difficulties students experience in

regard to academic performance. The sub-scale has a reported internal consistency of .87

(Anton and Reed, 1991).

The psychosocial variables that were measured in the questionnaire were

commitment to university, student-institution integration, satisfaction with university,

career orientation, financial difficulties, social support, and psychological health. These

questions were rated along a 4 point likert scale from “did not apply to me” (0) to “applied

to me very much or most of the time” (3). Most of the questions pertaining to student-

institution integration and commitment, social support, financial situation, and career

orientation were adapted from Himelstein’s (1992) questionnaire used to identify students

at risk of withdrawal and/or failure.

While the psychometric properties of the questionnaire were not reported,

Himelstein (1992) found student’s indicating the statements applied to them very much,

had higher rates of course completion and higher GPA’s than students who indicated low

applicability. The remainder of these questions were based on factors that previous

research has indicated are predictive of academic performance (Gerdes and Mallinckrodt,

1994; Lecompte, Kaufman and Rousseeuw, 1983; Pantages and Creedon, 1975; Wince and

Borden, 1995).

The psychosocial variable of psychological health was assessed using the short
 Factors predicting academic performance 16

version (21 items) of the Depression, Anxiety and Stress Scale (DASS) (Lovibond and

Lovibond, 1995). The depression sub-scale assesses lack of positive affect, low self

esteem, and feelings of hopelessness; the anxiety sub-scale assesses hyperarousal,

nervousness, and apprehension; and the stress sub-scale assesses feelings of frustration,

irritability, and tension (Brown, Chorpita, Korotitsch, and Barlow, 1997). The sub-scales

of the short version consist of 7 items for which participants rate themselves along a 4

point severity/frequency likert scale. Scores are summed, multiplied by two, and

compared to Australian norms to provide an indication of the degree of stress, anxiety, and

depression experienced by the participants. Internal consistency of the sub-scales of the

DASS is high (Cronbach’s alpha = .96, .89, and .93 for depression, anxiety, and stress

respectively) (Brown et al., 1997). Support has also been found for the discriminant

validity and temporal stability of the DASS (Brown et al., 1997).

Self-efficacy was assessed by measuring participants responses to the statement

“Based on my academic ability, I expect my grades will be above average” (Himelstein,

1992). The response was also scored along a 4 point likert scale from “did not apply to

me” (0) to “applied to me very much or most of the time” (3). According to self-efficacy

theory (Lent, Brown and Larkin, 1984), students scoring high on the scale exhibit high

self-efficacy and students scoring low on the scale exhibit low self-efficacy.

Attributional style was measured using the Attributional Style Questionnaire (ASQ)

developed by Peterson, Semmel, von Baeyer, Abramson, Metalsky and Seligman (1982).

This questionnaire provides scores on the dimensions of internality (internal vs external),

stability (stable vs unstable), and pervasiveness (global vs specific) of explanations for

good or bad events. It is a self report instrument on which participants give a cause for six

good and six bad hypothetical life events and then situate the cause along a 7 point scale

for three corresponding questions. These scores are combined to produce a composite
 Factors predicting academic performance 17

negative (CoNeg) and composite positive score (CoPos), with higher scores on CoNeg

indicating a more negative attributional style and higher scores on CoPos indicating a more

positive attributional style (range from 3 to 21). A composite overall score of positive and

negative scores is obtained by subtracting CoNeg from CoPos, with high scores indicating

an overall positive attributional orientation and low scores indicating an overall negative

attributional orientation (range from -18 to +18).

Peterson et al. (1982) found the ASQ to have acceptable levels of reliability for

CoPos and CoNeg (alpha = .75 and .72 respectively). Peterson and Seligman (1984)

reported the ASQ has high consistency of scores across tests, and high criterion related

validity, and that attributional style has high temporal stability.

The demographic variable of employment was obtained in the questionnaire. The

demographic variables of age, gender, and student workload, the academic variable of

university entry score, and first semester GPA's were obtained by accessing the Student

Information System (SIS), which is a computerised record of student information at the

University. Students whose university entry score were not available on the SIS system

were contacted by e-mail and asked to forward on their university entry score.

Procedure

The purpose of the study was explained to students in lecture time four to eight

weeks prior to the end of semester examinations. The voluntary nature of the study was

explained and students who chose to complete the questionnaire provided informed

consent to participate in the study. Students then proceeded to fill out the questionnaire,

which took approximately twenty-five minutes to complete. Questionnaires were collected

immediately after completion. The return rate was over 65% for the Science students but

there was a poor response rate for the Information Technology students.

Results
 Factors predicting academic performance 18

Descriptive Analysis

The means of all the variables used in the analysis are reported in Table 1.

Insert Table 1 Here

Academic Predictors of Academic Performance

A standard regression was performed between GPA as the dependent variable and

university entry score as the independent variable. University entry score was significantly

related to GPA. University entry score accounted for 39% of the variance in GPA.

Inspection of the beta values shows a negative relationship between university entry score

and GPA, such that as university entry score gets higher academically (closer to one), GPA

gets higher (closer to 7). Table 2 summarises the results of the regression analyses in this

study.

Insert Table 2 Here

A standard regression was performed between GPA and self reported study skills.

Self reported study skills was not a significant predictor of GPA.

Psychosocial Predictors of Academic Performance

A standard regression was performed between GPA and student institution

integration. Student institution integration was significantly related to GPA accounting for

3% of the variance in GPA. Inspection of the beta values shows a negative relationship

between student institution integration and GPA. A standard regression was performed

between GPA and commitment, GPA and satisfaction, and GPA and career orientation.

None of these variables were significant predictors of GPA.

A one way ANOVA was performed between GPA as the dependent variable and

level of financial difficulty as the independent variable. There was no significant

difference between GPA for levels of financial difficulty, F (3,171) = 2.46, p >.05.

The hypothesis concerning social support can not be analysed due to the extreme
 Factors predicting academic performance 19

skewness of the distribution. The majority of students reported high levels of social

support for their university studies.

Three standard regressions were performed between GPA and each of the variables,

depression, anxiety and stress. These three variables were not significant predictors of

GPA.

Cognitive Appraisal as a Predictor of Academic Performance

A standard regression was performed between GPA and self efficacy. Self efficacy

was significantly related to GPA accounting for 8% of the variance in GPA. Inspection of

the beta values shows a positive relationship between self efficacy and GPA.

A standard regression was performed between GPA and the ASQ dimension of

CPCN and no significant relationship was identified. A standard regression was performed

between GPA and the ASQ dimension of CoPos and no significant relationship was

detected. A standard regression was performed between GPA and the ASQ dimension of

CoNeg. CoNeg was a significant positive predictor of GPA at the .10 level accounting for

2% of the variance in GPA.

Demographic Predictors of Academic Performance

A one way ANOVA was performed between age and GPA. There was no

significant difference found across age groups, F (1,173) = .23, p = .63.

A one way ANOVA was performed between GPA and employment

responsibilities. A significant difference in GPA was found for students with different

employment responsibilities, F (2, 172) = 10.73, p < .0001. A post hoc analysis (Tukeys

HSD) revealed part time employed students have significantly poorer GPA’s than full time

employed or unemployed students, with no significant difference between the GPA's of

full time employed students and unemployed students. A one way ANOVA was

performed between GPA and student workload. No significant difference was found
 Factors predicting academic performance 20

between the GPA's of part time and full time students, F (1,173) = .11, p = .74. As small

numbers in some cells prevented a factorial ANOVA from being performed, a chi square

analysis was performed between employment and student workload. It appeared that full

time employees were often part time students, part time employees were often full time

students, and students with no employment responsibilities appeared to be full time

students (p < .0001). Considered together, full time employees with a part time student

workload and full time students with no employment commitments appear to have

significantly higher GPA’s than full time students who work part time.

A one way ANOVA was performed between GPA and gender. There was no significant

difference between males and female’s GPA's, F (1,173) = .17, p = .68.

Academic, Psychosocial, Cognitive, and Demographic Predictors of Academic

Performance

To inspect the data further, an hierarchical regression was performed between GPA

as the dependent variable and university entry score entered in the first block, with student

institution integration, self efficacy, and employment responsibilities entered in the second

block (See Table 3).

Insert Table 3 Here

Integration and self efficacy were found to contribute significantly to the prediction

of GPA over and above that accounted for by university entry score alone. The addition of

measures of integration and self efficacy improves the prediction of GPA to 51 %, a 12 %

improvement on the prediction of GPA based on OP alone.

Discussion

The main purpose of this study was to undertake a prospective investigation of the

relationship between academic, psychosocial, cognitive, and demographic factors and

academic achievement in university.

 Factors predicting academic performance 21

Academic Predictors of Academic Performance

As predicted, university entry scores were a significant predictor of student's GPA's

at the end of the first semester of their course of study. Students with high university entry

scores were likely to continue this high academic achievement in university. However,

university entry scores need to be interpreted with caution, as they explain less than half of

the variance in GPA. Unfortunately the prediction of GPA from university entry scores for

different age groups could not be examined due to insufficient background information

recorded for mature age students.

The hypothesis that study skills were related to GPA was not supported.

Self reported study skills were not significantly different for different levels of academic

achievement. An explanation for this is that students may hold unrealistic views of their

study skills or report a level of study skill deemed acceptable for university. Students with

poorer study skills may not be aware of how much of a problem they have with studying at

the university level, or may be aware of problems but be unwilling to report such

difficulties. Alternatively, students with well-developed study skills may not necessarily

see themselves as particularly gifted in their study skills at the university level, or may

report a lower, more conservative level of study skill.

Psychosocial Predictors of Academic Performance

As predicted, student institution integration was a significant predictor of academic

performance, but interestingly, the relationship was a negative correlation. That is,

students who indicated high levels of integration into university tended to have poorer

GPA’s than students indicating low levels of integration.

The finding that integration has an adverse affect on academic achievement is

contrary to Tinto’s (1975) model of integration that suggests integration is crucial for

positive outcomes at university. While Tinto (1975) recognised that certain social groups
 Factors predicting academic performance 22

may adversely affect the grades of students who are members of those groups, this caveat

was ignored in the development of his model which was generalised to cover all students.

However, McInnes, James and McNaught (1995) suggest that integration may have an

adverse affect on academic achievement in some student cultures and may be unnecessary

for certain students.

There are several explanations for these research findings. Firstly, the distinction

between attrition and academic achievement must be noted. Tinto’s model was originally

developed to explain student attrition, and while researchers have found support for this

model in relation to academic achievement, it is possible that differences arise from the

distinction between the two concepts. High academic achievement is not necessarily

related to retention and poor academic performance does not always result in attrition.

Thus, it is possible that while student institution integration has a negative correlation with

academic achievement, integration may still be positively correlated with student retention.

Secondly, the differences in findings as reported in this study and previous research

based on Tinto's model may have arisen due to cultural differences between Australian and

American samples. The majority of the research based on Tinto's model has been

undertaken in America. It is arguable that the American student cultural context might be

different from the Australian context. While integration into university may be important

for the academic performance of American students, it appears that a lack of integration

does not adversely affect the academic performance of Australian students, but may even

contribute to positive results.

Finally, with changes in technology and university policies, the characteristic well-

integrated student identified in previous research may have changed. With the advent of

the Internet and e-mail, the social nature of universities may be changing, and studying in

isolation may have become adaptive for a sub-group of high achieving students. As
 Factors predicting academic performance 23

McInnes, James and McNaught (1995) state: “There is a likelihood that student identity, as

we know it, is declining. In addition to the influence of flexibility and openness in course

structures and delivery, there are broader social forces involved” (p. 88).

The hypothesis that commitment to university would be a significant predictor of

academic achievement was not supported. The majority of students indicated a moderate

to high level of commitment to university. That is, the majority of students indicated they

would reduce outside responsibilities that interfered with university study, they believed

attending class was important, and were confident of completing their course of study. It is

possible that the slight negative skewness of the sample affected the significance of

commitment as a predictor of GPA.

The hypothesis that career orientation would be predictive of GPA was not

supported in this sample of first year students. Students with different levels of clarity of

career goals did not differ significantly in their attained GPA. The majority of students

indicated moderate to high levels of clarity in their career orientation. One of the possible

explanations for this result is that career orientation fluctuates and affects GPA differently

at different times in the student’s university career. Students start their university year

with a perception of what the university course will offer in terms of career choices. With

new experiences and changing perceptions of the university course, the reality of possible

careers may change. Students may change their career goals in a way which may no longer

fit with their course of study, and previous research would suggest that such an

incongruence would affect student’s academic performance at this time (Himelstein, 1992;

McInnes, James, and McNaught, 1995).

Psychological health was not a significant predictor of academic achievement.

However, as psychological health was measured in the middle of semester, it is possible

that depression, anxiety and stress levels were already affected by university factors
 Factors predicting academic performance 24

(exams and assignments). The non-significant results obtained in measures of

psychological health may be due to the time of semester that results were obtained.

Szulecka, Springett and dePauw’s (1987) research on psychological health was obtained

prior to commencement of the university year, which they stated was important so that

students were “still strongly influenced by pre-university experiences and to be least

contaminated by university-related factors” (p. 83). It might be argued that measuring

student’s psychological health in the middle of semester may have obtained results that

were influenced by these extraneous variables, and may not have identified the student’s

characteristic level of psychological functioning. Another argument is that, rather than

having a direct relationship with GPA, psychological health may influence other predictors

of academic performance, such as student institution integration, satisfaction with

university and self efficacy.

Cognitive Appraisal as a Predictor of Academic Performance

The hypothesis that self efficacy would be positively related to academic

performance was supported. Students reporting high self efficacy of achieving above

average grades had significantly higher GPA’s than students reporting low self efficacy of

achievement. This finding is in line with previous research in the field (Lent, Brown and

Larkin, 1984, 1987).

The hypothesis that an optimistic attributional style would be related to higher

academic achievement was not supported. Interestingly, the opposite relationship was

found, that is, a pessimistic attributional style was predictive of higher GPA’s at the .10

probability level.

Demographic Predictors of Academic Performance

The hypothesis concerning the influence of employment responsibilities and

student workload was partially supported. Full-time students with no employment

 Factors predicting academic performance 25

responsibilities appeared to have higher GPA’s than full-time students with part-time

employment responsibilities. In addition, it was found that part-time students with full-

time employment responsibilities had significantly higher GPA’s than full-time students

with part-time employment responsibilities. The poorest GPA’s were identified amongst

students with full-time study commitments and part-time employment.

The difference in GPA between full-time students with no employment and with

part-time employment can possibly be explained by time restraints. While full-time

students with no employment can devote their time to study for university, part-time

employment limits the amount of time available to devote to study. The difference in

GPA’s between full-time students with part-time employment and part-time students with

full-time employment is somewhat more complex and obscure. It is possible that part-time

students with full-time employment are highly motivated to study and have clear career

goals. They may also have well developed time management skills as a product of their

full-time employment responsibilities, which might benefit them in their university studies.

Full-time students with part-time employment may not have developed these time

management skills and career goals, and lacking these skills and goals may adversely

affect their academic performance.

Academic, Psychosocial, Cognitive and Demographic Predictors of Academic

Performance

Student institution integration and self efficacy were significant predictors of GPA

after accounting for the differences in OP scores between students. This model improved

the prediction of GPA from OP score by 12%.

We have several recommendations for future research. From a methodological

perspective, adequate recording of university entry scores/high school grades of all

students is important. Implementing stringent record keeping procedures at the university

 Factors predicting academic performance 26

level would enable researchers to fully examine the relationship between age, previous

academic performance and university achievement.

Attributional style research is a relatively new field and several research

possibilities have been highlighted in this study. The attributional style of students across

fields of study could be examined to identify the characteristic attributional styles of

groups of students. The behaviour invoked by optimistic and pessimistic students in

response to negative events may be investigated to gain a better understanding of the

complex relationship between attributional style and academic performance. The

relationship between self efficacy, attributional style, and academic performance could be

examined to identify the influence of a student’s self efficacy on their attributions for

events, and the corresponding influence on their academic achievement. Such studies may

also be extended across cultures to examine the potential 'culture-bound' nature of

attributional style.

We suggest several important implications for student support interventions and

curriculum change. In general, higher university entry scores and high self efficacy are

related to higher academic achievement at university. In terms of university entry scores,

universities need to make realistic appraisals of the academic demands of particular

courses and set the entry score for school leavers at a level based on academic challenge

rather than on course demand. As high university entry scores are moderately correlated

with high GPA’s, entry to courses with a high level of difficulty should be set at a realistic

level to avoid undue problems for both students with lower university entry scores and for

the university. Alternatively, specialised enhancement programs need to be introduced and

evaluated to provide students with additional skills.

In terms of self efficacy, confidence building programs could be targeted to at-risk

students and implemented at the start of first semester. These programs would be aimed at
 Factors predicting academic performance 27

improving student’s self efficacy toward academic achievement and could include an

introduction to the style of work expected at university, enhancement of study skills,

orientation to different university resources and services, and general confidence raising

activities.

Predictors of academic performance influence specific groups at different levels.

School leavers, both males and females, are negatively influenced by their degree of

integration into the university. There are two suggestions for interventions/curriculum

changes to improve school leaver's GPA’s. Firstly, individual work and individual

assessment should be encouraged for school leavers. Rather than trying to encourage

group work based on the assumption that integration will improve academic performance,

the benefits of individual study may be promoted. Encouraging individual work may

decrease the likelihood that school leavers will be influenced by the possible disinclination

toward academic achievement identified in this study.

Alternatively, interventions could be aimed at promoting academic achievement as

an important part of integration into the university. University orientation weeks, with

their often heavy emphasis on social activities, having fun, and alcohol consumption, may

in fact be promoting this disinclination toward academic achievement. It may be necessary

to rethink the activities promoted in orientation week and put an increased emphasis on

study skills and academic achievement as integral parts of university life. Promotion of

study groups (as opposed to social groups) and high academic achievement as socially

acceptable and encouraged may help to change this negative view of academic

achievement that appears to be promoted in orientation week.

This study has addressed the issue of student achievement by examining

prospectively the relationship between academic, psychosocial, cognitive, and

demographic variables and the academic performance of first year Australian university
 Factors predicting academic performance 28

students at an urban commuter-based university. While previous academic performance

was the most significant predictor of academic achievement in university, several other

factors were identified that influence university grades. Measuring a broad range of

academic, psychosocial, cognitive, and demographic variables led to the development of a

model which improved the prediction of academic achievement from a prediction based on

previous academic performance alone.

Identification of the factors that influence academic performance is important at

this point in the history of Australian higher education. With the expanding of Australian

universities over the last decade to provide equal access for all, comes an increasing

diversity of student's characteristics and needs. To fully embrace this equity initiative,

universities must cater for this diverse student population and implement strategies and

interventions based on sound research, to give all students a fair chance for academic

success.

References

Abbott-Chapman, J., Hughes, P., & Wyld, C. (1992). Monitoring student progress:

A framework for improving student performance and reducing attrition in higher

education. Hobart: National Clearinghouse for Youth Studies.

Abramson, L. Y., Seligman, M. E. P., & Teasdale, J. D. (1978). Learned

helplessness in humans: Critique and reformulation. Journal of Abnormal Psychology,

87(1), 49-74.

Anton, W. D., & Reed, J. R. (1991). College Adjustment Scales: Professional

 Factors predicting academic performance 29

Manual. USA: Psychological Assessment Resources.

Baker, R.W., & Siryk, B. (1984). Measuring adjustment to college. Journal of

Counselling Psychology, 31 (2), 179-189.

Braxton, J. M., Brier, E. M., & Hossler, D. (1988). The influence of student

problems on student withdrawal decisions: An autopsy on "Autopsy" studies. Research in

Higher Education, 28(3), 241-253.

Brown, T.A., Chorpita, B.F., Korotitsch, W., & Barlow, D.H. (1997). Psychometric

properties of the Depression Anxiety Stress Scales (DASS) in clinical samples.

Behavioural Research and Therapy, 35 (1), 79-89.

Cabrera, A. F., Nora, A., & Castaneda, M. B. (1993). College persistence:

Structural equations modelling test of an integrated model of student retention. Journal of

Higher Education, 64(2), 123-137.

Chapman, D.W., & Pascarella, E.T. (1983). Predictors of academic and social

integration of college students. Research in Higher Education, 19 (3), 295-322.

Clark, E. E., & Ramsay, W. (1990). Problems of retention in tertiary education.

Education Research and Perspectives, 17(2), 47-57.

Cohen, J., & Cohen, P. (1975). Applied multiple regression/correlation analysis for

the behavioural sciences. New York: Erlbaum.

Cone, A. L., & Owens, S. (1991). Academic and locus of control enhancement in a

freshman study skills and college adjustment course. Psychological reports, 68, 1211-1217.

Everett, J.E., & Robins, J. (1991). Tertiary entrance predictors of first year

university performance. Australian Journal of Education, 35 (1) 24-40.

Gerdes, H., & Mallinckrodt, B. (1994). Emotional, social, and academic adjustment

of college students: A longitudinal study of retention. Journal of Counselling and

Development, 72, 281-288.

 Factors predicting academic performance 30

Himelstein, H. C. (1992). Early identification of high-risk students: Using non-

cognitive indicators. Journal of College Student Development, 33, 89-90.

Houston, H.K. (1971). Sources, effects, and individual vulnerability of

psychological problems for college students. Journal of Counselling Psychology, 18 (2),

157-165.

Lecompte, D., Kaufman, L., & Rousseeuw, P. (1983). Search for the relationship

between interrupted university attendance of first year students and some psychosocial

factors. Acta Psychiatrica Belgica, 83, 609-617.

Lecompte, D., Kaufman, L., Rousseeuw, P., & Tassin, A. (1983). Search for the

relationship between academic performance and some psychosocial factors: The use of a

structured interview. Acta Psychiatrica Belgica, 83, 598-608.

Lent, R.W., Brown, S.D., & Larkin, K.C. (1984). Relation of self-efficacy

expectations to academic achievement and persistence. Journal of Counselling Psychology,

31 (3), 356-362.

Lent, R. W., Brown, S. D., & Larkin, K. C. (1987). Comparison of three

theoretically derived variables in predicting career and academic behaviour: Self efficacy,

interest congruence, and consequence thinking. Journal of Counselling Psychology, 34(3),

293-298.

Lovibond, S.L., & Lovibond, P.F. (1995). The structure of negative emotional

states: Comparison of the Depression Anxiety Stress Scales (DASS) with the Beck

Depression and Anxiety Inventories. Behaviour Research and Therapy, 33 (33), 335 – 343.

McInnes, C., James, R., & McNaught, C. (1995). Diversity in the initial

experiences of Australian undergraduates. Committee for the Advancement of University

Teaching (On-line). Available: www.online.anu.edu.au/caut/com…e/FYEfront.html .

McKeown, B., Macdonell, A., & Bowman, C. (1993). The point of view of the
 Factors predicting academic performance 31

student in attrition research. The Canadian Journal of Higher Education, 23, 65-85.

Mutter, P. (1992). Tinto's theory of departure and community college student

persistence. Journal of College Student Development, 33, 310-317.

Organisation for Economic Cooperation and Development. (1997). Thematic

review of the first years of tertiary education in Australia. Department of Employment,

Education, Training, and Youth Affairs (On-line). Available: www.deetya.gov.au/

divisions/hed/operations/theme.htm .

Pantages, T.J., & Creedon, C.F. (1975). Studies of college attrition: 1950-1975.

Review of Educational Research, 48 (1), 49-101.

Pascarella, E. T., Duby, P. B., Miller, V. A., & Rasher, S. P. (1981). Preenrollment

variables and academic performance as predictors of freshman year persistence, early

withdrawal, and stopout behaviour in an urban, nonresidential university. Research in

Higher Education, 15(4), 329-347.

Pascoe, R., McClelland, A., & McGaw, B. (1997). Perspectives on selection

methods for entry into higher education in Australia. AGPS: Canberra.

Peterson, C., & Barrett, L. (1987). Explanatory style and academic performance

among university freshman. Journal of Personality and Social Psychology, 53(3), 603-607.

Peterson, C., & Seligman, M. E. P. (1984). Causal explanation as a risk factor for

depression: Theory and evidence. Psychological Review, 91(3), 347-374.

Peterson, C., Semmel, A., von Baeyer, C., Abramson, L. Y., Metalsky, G. I., &

Seligman, M. E. P. (1982). The attributional style questionnaire. Cognitive Therapy and

Research, 6, 287-299.

Pierce, M. A., & Henry, J. W. (1993). Attributional style as a predictor of success

in college mathematics (ERIC Document Reproduction Services No. ED 365 528).

Statesboro, GA: Georgia Southern University.

 Factors predicting academic performance 32

Power, C., Robertson, F., & Baker, M. (1987). Success in higher education.

Canberra: Australian Government Publishing Service.

Rickinson, B., & Rutherford, D. (1995). Increasing undergraduate student retention

rates. British Journal of Guidance and Counselling, 23 (2), 161-171.

Rickinson, B., & Rutherford, D. (1996). Systematic monitoring of the adjustment to

university of undergraduates: A strategy for reducing withdrawal rates. British Journal of

Guidance and Counselling, 24(2), 213-225.

Rosenman, L. (1996). The broadening of university education: An analysis of entry

restructuring and curriculum change options. Department of Employment, Education,

Training, and Youth Affairs (On-line). Available: www.deetya.gov.au/divisions/hed/

operations/year13.htm .

Seligman, M. E. P. (1991). Learned Optimism. Milsens Point, NSW: Random

House.

Seligman, M.E.P. (1998). Building human strength: psychology's forgotten

mission. Monitor: American Psychological Association, 2.

Seligman, M. E. P., & Schulman, P. (1986). Explanatory style as a predictor of

productivity and quitting among life insurance sales agents. Journal of Personality and

Social Psychology, 50(4), 832-838.

Spady, W. G. (1971). Dropouts form higher education: Toward an empirical model.

Interchange, 2(3), 38-62.

Stanley, G (1997). Perspectives on selection methods for entry into higher

education in Australia. Department of Employment, Education, Training, and Youth

Affairs (On-line). Available: www.deetya.gov.au/nbeet/hec/publicat/cr58/methaus.htm .

Szulecka, T. K., Springett, N. R., & de Pauw, K. W. (1987). General health,

psychiatric vulnerability and withdrawal from university in first-year undergraduates.

 Factors predicting academic performance 33

British Journal of Guidance and Counselling, 15(1), 82-89.

Terenzini, P. T., & Pascarella, E. T. (1978). The relation of students' precollege

characteristics and freshman experience to voluntary attrition. Research in Higher

Education, 9, 347-366.

Tinto, V. (1975). Dropout from higher education: A theoretical synthesis of recent

research. Review of Educational Research, 45(1), 89-125.

Tominey, M. F. (1996). Attributional style as a predictor of academic success for

students with Learning Disabilities and/or Attention Deficit Disorder in post secondary

education (ERIC Document Reproduction Services No. ED 407 815). Austin: University

of Texas.

Tracey, T.J., & Sedlacek, W.E. (1982). Non-cognitive variables in predicting

academic success by race (ERIC Document Reproduction Services No. ED 219012).

Maryland: University of Maryland.

Weiner, B. (1985). Attributional theory of achievement motivation and emotion.

Psychological Review, 92(4), 548-573.

West, R. (1998). Learning for life: Higher education review - Final report.

Department of Employment, Education, Training, and Youth Affairs (On-line). Available:

www.deetya.gov.au/divisions/hed/hereview/execsum1.htm .

Wilson, T. D., & Linville, P. W. (1982). Improving the academic performance of

college freshman: Attribution therapy revisited. Journal of Personality and Social

Psychology, 42(2), 367-376.

Wilson, T. D., & Linville, P. W. (1985). Improving the performance of college

freshman with attributional techniques. Journal of Personality and Social Psychology,

49(1), 287-293.

Wince, M. H., & Borden, V. M. H. (1995). When does student satisfaction matter?
 Factors predicting academic performance 34

(ERIC Document Reproduction Services No. ED 386 990). Massachusetts: Association for

Institutional Research.

Table 1

Descriptives for Data in Study

Variable Mean SD

University entry score 5.91 2.69

Study skills 1.59 .76

Integration 1.62 .64

Commitment 2.23 .54

Satisfaction 2.1 .54

Career orientation 1.83 .94

Social support 2.70 .71

Depression 2.86 1.49

 Factors predicting academic performance 35

Anxiety 2.38 1.32

Stress 3.40 1.39

Self efficacy 1.87 .92

CPCN 2.82 2.56

CoNeg 12.36 1.87

CoPos 15.18 1.83

Table 2

Summary of Regression Analyses

VARIABLE B SE B 

Uni entry score -.26 .03 -.62****

Study skills .15 .11 .10

Integration -.32 .13 -.18**

Commitment .20 .16 .09

Satisfaction .29 .16 .14*

Career orientation -.01 .09 -.01

Depression .04 .06 .05

Anxiety .02 .07 .03

Stress .09 .06 .11

 Factors predicting academic performance 36

Self efficacy .34 .09 .28***

CPCN -.05 .03 -.12

CoNeg .08 .05 .13*

CoPos -.02 .05 -.03

**** p < .0001. *** p < .001. ** p < .05. * p < .10

Table 3

Psychosocial, Cognitive and Demographic Predictors of GPA Removing OP

VARIABLE B SE B  R SQUARE

Step 1

OP -.26 .03 -.62****

Total R Square .39****

Step 2

Integration -.55 .14 -.29***

Self efficacy .38 .10 .30***

Employment -.24 .15 -.11

R Square Change .12****

Total R Square .51****

**** p < .0001. *** p < .001. ** p < .05. * p < .10
Factors predicting academic performance 37