An Examination Of The Relationship Between Contextual And Technical Performance In An Academic Setting

Arthur Poropat

Griffith University

(E-mail: )

Paper presented at the ESRC Teaching and Learning Research Programme, First Annual Conference - University of Leicester, November 2000

ABSTRACT

Job Performance is increasingly conceived of as including two components: technical performance reflecting job requirements and contextual performance which covers non-prescribed activities such as team-work and support. Within educational settings performance is still typically tied to assessment measures such as marks which reflect technical performance. This study therefore examines the relationship between contextual and technical performance within an educational setting. Participants were students enrolled within an introductory management subject. In all 205 students both agreed to participate and provided useable data for this study. Two measures of Contextual Performance were used. The Computerised Adaptive Rating Scales (CARS) and attendance at tutorials. Technical performance was measured by marks on two assignments. A factor scale labelled active support was developed on the basis of CARS. Active support and attendance both predicted performance on the Essay (0.24 and 0.35 respectively; combined R2 = 0.120) and the Take-Home Exam (0.15 and 0.32 respectively; combined R2 = 0.085). The significant role of contextual performance demonstrated implies that methods of teaching should incorporate socially facilitative aspects as well as more traditional measures. University teachers can actively promote increased contextual performance through teaching methods which encourage value alignment. Further exploration of the links between educational and employment related contextual performance may be of both pedagogical and organisational value.

INTRODUCTION

Performance is conceived of in a variety of ways in research but it is a key variable within both organisational and educational settings. Despite this, performance as a construct has received comparatively little theoretical attention when compared with other variables relevant to these areas, possibly because researchers tend to emphasise independent variables (eg individual differences, treatment conditions) or because performance tends to be controlled by people other than the researcher (Campbell, McCloy, Oppler & Sager, 1993).

Performance has often been operationalised by measures over which an individual has limited control (eg productivity and efficiency confound organisational constraints) and that therefore confuse the relationship between predictor variables and criteria. However performance is more usefully defined as actual behaviour that can be scaled and measured in terms of proficiency rather than outcomes (Campbell, et al, 1993). Ratings of behaviour correspond well with this definition and in educational settings ratings such as grades and marks reflect actual performance to the extent that students and trainees are not subjected to arbitrary or biased assessment.

Recent models of job performance have tended to move from an assumption that performance is unitary (a behavioural equivalent to the g factor in IQ) to models which emphasise separate aspects of performance (Borman, Hanson & Hedge, 1997). Campbell (1990) and Campbell, Gasser and Oswald (1996) present a multi-factor model of job performance based on a review of the job performance literature and provide confirmatory research from military settings. The eight components of job performance identified in their model are:

  1. Job-Specific Task Proficiency
  2. Non-Job-Specific Task Proficiency
  3. Written And Oral Communication Task Proficiency
  4. Demonstration Of Effort
  5. Maintenance Of Personal Discipline
  6. Facilitation Of Peer And Team Performance
  7. Supervision/Leadership
  8. Management/Administration

Campbell, et al (1996) deny that these components represent orthogonal factors of job performance and also state that not all of these components are necessarily present in every job nor that they are the last word in defining the performance domain. However they do suggest that these components account for most of the variation in job performance assessments as well as presenting evidence which suggests the relative independence of the various components and their relevance for performance research. One of the claims made by Campbell et al (1993) is that this model of job performance appears to be equally applicable to educational, training and laboratory task performance. In each situation people are required to perform to certain levels in a dynamic, socially demanding environment.

An alternative and simpler approach to identifying the various aspects of performance grew out of path analytic studies which suggested the presence of several key components. Borman & Motowidlo (1993) in particular developed a model based on two key types of performance - technical and contextual - which are somewhat reminiscent of the venerable distinction between task and maintenance processes in group research. Technical performance relates to what Borman & Motowidlo (1993) call the 'technical core' of the organisation or job, the activities directly or indirectly involved with transforming resources into products for economic exchange. Technical task activities vary considerably from job to job, are dependent on knowledge, skills and abilities, and are role-prescribed.

Contextual performance on the other hand involves activities directed at maintaining the inter-personal and psychological environment that needs to exist to allow the technical core to operate. As such contextual activities are common to most if not all jobs, are dependent upon motivations and predispositions such as personality, and are rarely role-prescribed. Examples of contextual performance are activities such as:

"Volunteering to carry out task activities that are not formally part of the job

Persisting with extra enthusiasm or effort when necessary to complete own task activities successfully

Helping and cooperating with others

Following organizational rules and procedures even when personally inconvenient

Endorsing, supporting and defending organizational objectives" (Borman & Motowidlo, 1993, p73).

Borman and Motowidlo's model is simpler but broadly comparable with Campbell et al's (1996). Components 1, 2 3, and possibly component 8 of Campbell et al's (1996) model are linked to technical performance, while components 4, 5, 6 and possibly 7 appear to be examples of contextual performance. It should be noted that the proponents of both these models freely acknowledge the strengths of the alternative models and the limitations of their own (Campbell, Gasser & Oswald, 1996; Borman & Motowidlo, 1997). Despite their different origins these should be seen as complementary models rather than competitive ones.

Other writers have also looked at alternatives to traditional understandings of performance with concepts that are directly comparable with the technical/contextual distinction. Organisational citizenship behaviour (Smith, Organ & Near, 1983) was identified as the contributions made by employees to their fellows and the organisation overall which extended beyond the requirements of formal roles or job descriptions. As such organisational citizenship behaviour has been explicitly linked to contextual performance (Borman & Motowidlo, 1993; Organ & Ryan, 1995).

Contextual performance and similar variables have been growing in importance in discussions of organisational performance criteria (Borman, Hanson & Hedge, 1997). In part this is because variables like contextual performance have been shown to have roughly equal influence with technical performance on overall performance ratings by supervisors (Motowidlo & Van Scotter, 1994; Borman, White & Dorsey, 1995). Podsakoff & MacKenzie (1997) review evidence that the impact of contextual variables on technical performance has economically significant implications, predicting between 15% and 43% of variance on such measures as sales, production and quality. Contextual performance has also grown in theoretical importance as a mediating variable between personality and general ratings of performance (Konovsky & Organ, 1996; Borman, Hanson & Hedge, 1997; Borman & Motowidlo, 1997; Motowidlo, Borman, & Schmit, 1997).

Discussion of performance within educational settings has not had the same discussion of types of performance as that reviewed so far for organisational settings. Mostly performance has been discussed from the perspective of how best to assess it with the nature of performance itself being left largely assumed or explicitly tied to educational objectives (cf. Palomba & Bamba, 1999). When performance is discussed by writers in tertiary education this is typically done by emphasising its active and 'real world' nature and contrasting performance with the knowledge and/or skills which underlie it (Moss, 1992; Biggs, 1999). The multi-variate model of performance described above has not as yet been investigated or applied to educational settings. This study therefore aims to discover the nature of the relationship between contextual and technical performance within an educational setting.

METHOD

Sample

Participants were students enrolled within an introductory management subject. All measures used in the study were conducted as part of the teaching or assessment process, but inclusion within the analysis for this research was voluntary: students were informed that participation would have no positive or negative influence on their marks and analyses were conducted after finalisation of marks. In all 205 students both agreed to participate and provided useable data for this study.

Measures

Contextual Performance. The measurement of contextual performance is a complex and developing field, with much work still underway attempting to identify the basic dimensionality of the construct (Motowidlo, Borman, & Schmit, 1997). Two measures of contextual performance were used in this study. The Computerised Adaptive Rating Scales or CARS (Borman, 1999) were developed using the framework established by Motowidlo, Borman and Schmit (1997). This instrument attempts to measure three dimensions of contextual performance – Personal Support, Organisational Support and Conscientious Initiative. Additionally attendance at tutorials was recorded and used as a measure of contextual performance in line with previous research (eg. Hattrup, O’Connell & Wingate, 1998).

Technical Performance. The measures of technical performance used within this study were the marks for two pieces of assessment completed within the subject. These pieces were an individual essay, and a take-home exam consisting of three short essays. The assessors used standardised criteria for assessing individual pieces of assessment and the participants were provided with these criteria in advance as part of the learning process (Nightingale, Wiata, Toohey, Ryan, Hughes & Magin, 1996). The criteria used were use of literature from relevant content area, logical argument and presentation, which correspond broadly with Campbell, Gasser and Oswald's (1996) first three components of job performance (Job-Specific Task Proficiency, Non-Job-Specific Task Proficiency, Written And Oral Communication Task Proficiency). An overall rating or mark was given to each assessment item on the basis of these criteria.

Procedure

All assessments were conducted as part of the learning process for the subject. The CARS scales were scored by fellow students after completion of a group project. Attendance was recorded by tutors at each tutorial and ratings of technical performance were conducted as assessment items were completed.

RESULTS

Contextual Performance

The CARS scales were relatively low on internal reliability as assessed by Cronbach Alpha (0.54 to 0.77) as well as being low on inter-rater reliability (0.47 to 0.60). Equally concerning is the fact that the three scales are highly inter-related, with their inter-correlations ranging from 0.62 to 0.78.

A principal components analysis and varimax rotation of the items in these scales revealed that the first factorial component accounted for 42% of the variance. On the basis of this a new scale consisting of eight of the original 12 items was constructed by eliminating all items with either low loadings on this first component or high loadings on multiple components. This scale was labelled active support and had both improved internal reliability (Cronbach Alpha = 0.86) and inter-rater reliability (r = 0.66) when compared with the original scales.

Technical Performance

Inter-rater reliabilities for assessments were computed using double-blind ratings on marks for both the essay and the take-home exam. The average inter-rater reliability was 0.85.

Statistical Relationships

The correlations between the various measures of contextual and technical performance are reported in Table 1. All measures of both contextual and technical performance were correlated in predicted directions.

Essay
Exam / 0.52**
Attendance / 0.35** / 0.32**
Active Support / 0.24** / 0.15* / 0.18**
Essay / Exam / Attendance

N = 205. * P < 0.05; ** P < 0.01, : P is one-tailed.

Table 1: Correlations between measures of contextual and technical performance.

Table 2 reports the results for the multiple regression of contextual performance measures on the essay, while Table 3 reports the multiple regression of these measures on the take-home exam. From these analyses it is apparent that both measures of contextual performance contribute to the statistical prediction of marks on both the essay and the exam, and that both measures provide a statistically significant independent contribution to the prediction of the respective criterion variables.

Beta / Partial R / Multiple R / R2
Attendance / .255** / .261** / .346** / .120**
Active Support / .216** / .224**

N = 205. * P < 0.05; ** P < 0.01, : P is one-tailed.

Table 2: Multiple regression of Attendance & Active Support on Essay.

Beta / Partial R / Multiple R / R2
Attendance / .252** / .254** / .291* / .085*
Active Support / .127* / .131*

N = 205. * P < 0.05; ** P < 0.01, : P is one-tailed.

Table 3: Multiple regression of Attendance & Active Support on Exam.

DISCUSSION

This study showed that contextual performance has a significant relationship with technical performance within an educational setting. Both measures of contextual performance provide independent contributions to the prediction of the two measures of technical performance. This relationship is similar to the link observed between contextual and technical performance within employment settings. Non-prescribed activities are related to the effective performance of prescribed activities in employment, and appear to be similarly related in educational settings.

One of the possible explanations for the relationship between contextual performance and technical performance in this study is that engaging in contextual behaviour would make it more likely that students would have opportunities to learn. In other words attendance may increase technical performance by increasing the exposure to the learning environment. Students who participate more increase their chances of receiving instruction and guidance, and it would be a great irony if all the effort put into teaching produced no impact upon student performance.

However the fact that active support correlated with technical performance independently of attendance suggests that there is more to this relationship than participation in learning activities. It appears that for some reason people who are more likely to assist others through actively supporting their efforts are enhancing their own efforts at the same time: a case of good behaviour providing its own reward. The fact that the personality variable of conscientiousness has been shown elsewhere to be positively correlated with contextual performance (Motowidlo, Borman, & Schmit, 1997) suggests that the relationship between contextual and technical performance may reflect a general willingness to make greater efforts. There is a need for experimental work to explore the causal nature of this link further.

Apart from research this study has implications for teaching practice. The significant role of contextual performance demonstrated here implies that methods of teaching and assessment should focus on socially facilitative aspects of performance as well as more traditional measures. Not only would this reflect growing trends in organisations (Hattrup et al, 1998), it also has potential to enhance the success of students within educational settings.

It may be tempting to extend assessment to cover contextual as well as technical performance along the lines of writers who suggest that students do what gets marked (Nightingale et al, 1996). However to the extent that marks have a similar impact to other extrinsic rewards such as money, this may be counter-productive in the long-run for many students. Although workers who are potentially able to achieve an extrinsic reward from contextual performance will increase their supportive behaviours, they are likely to drop these behaviours whenever the reward structure changes (Hogan, Rybicki, Motowidlo & Borman, 1998). However workers whose values are aligned with organisational values provide high levels of contextual performance regardless of extrinsic reward (Deckop, Mangel, & Cirka, 1999). Likewise workers who take pride in their tasks also display higher levels of contextual performance (Hodson, 1998).

Consequently it is worth considering educational approaches which are better at aligning student and faculty values such as student-centred learning (Biggs, 1999). This runs somewhat counter to the growing popularity of competency-based approaches to education (Hall, 1994; McKenzie, Mitchell, & Oliver, 1995; Smith, 1992). Competency-based-training is predicated on the assumption that what matters in life generally and work specifically are clearly defined skills and abilities, and that therefore training and education should likewise focus on these. Yet contextual performance is by its nature extra-role behaviour (Smith et al, 1983) and consequently difficult to specify and incorporate in clear competency requirements. Recognition of contextual performance does not negate the value of competency-based approaches, but does encourage a broader perspective on education and training.

In conclusion attention to the multi-dimensional nature of performance within educational settings is justified by this research. Further research on the nature of the relationship is likely to enhance understanding of the processes which lead to success in learning. However the findings presented here provide support for a greater emphasis on modelling and encouraging support and team-work among staff and students. This is likely to lead to better outcomes for students, a more pleasant environment for staff, and better-educated graduates for future employers.

REFERENCES

Biggs, J. (1999). Teaching for Quality Learning at University. Buckingham, U. K.: Society for Research into Higher Education and Open University Press.

Borman, W. (1999). Computerised Adaptive Rating Scales. Personal Communication.

Borman, W. C., & Motowidlo, S. J. (1993). Expanding the Criterion Domain to Include Elements of Contextual Performance. In N. Schmidt, W. C. Borman, A. Howard, A. Kraut, D. Ilgen, B. Schneider, & S. Zedeck (Eds.), Personnel Selection in Organizations (pp 71-98). San Francisco: Jossey-Bass.

Borman, W. C., & Motowidlo, S. J. (1997). Task performance and contextual performance: The meaning for personnel selection research. Human Performance. 10, 99-109.

Borman, W. C., Hanson, M. A., & Hedge, J. W. (1997). Personnel Selection. Annual Review of Psychology, 48, 299-337.