Internat. Jnl. for Educational and Vocational Guidance 3: 205–221, 2003. © 2003 Kluwer Academic Publishers. Printed in the Netherlands.

205

Factors Influencing Job Choice JAMES A. ATHANASOU Faculty of Education, University of Technology, P.O. Box 123, Broadway 2007, Sydney, Australia (E-mail: [email protected]) Received: July 2003; accepted: September 2003 Abstract. This research sets out a Perceptual-Judgemental-Reinforcement approach to job choice under conditions of complexity and uncertainty. It investigates the claim that job choices are based on seven implicit factors: such as the specific size of the occupation, the proportion of employees working full-time, the earnings, the job prospects, gender dominance in an occupation, the level of unemployment in the occupation and the predominant age group in the job. Nine case studies involving choices from 25 randomly selected advertised jobs are presented. Results indicated substantial idiosyncrasy in job choices. An individual logistic regression analysis indicated no statistically significant influence of key labour market indicators in any of the nine case studies. It was concluded that job choice was idiosyncratic; that individuals lacked insight into their job choices and probably relied upon relatively few unstated cues. The findings have direct implications for the relevance of occupational information and for key issues in the delivery of vocational guidance. Résumé. Facteurs influençant le choix professionnel. La présente recherche applique l’approche Perception-Jugement-Renforcement au choix professionnel dans des conditions de complexité et d’incertitude. Il soumet à examen la proposition selon laquelle les choix professionnels sont fondés sur sept facteurs implicites, à savoir l’ampleur spécifique du métier, la proportion des employés travaillant à temps plein, les salaires, les perspectives professionnelles, la suprématie de genre dans une profession, le niveau de sous-emploi dans une profession et le groupe d’âge prédominant dans le métier. Nous présentons neuf études de cas concernant le choix de 25 métiers aléatoirement sélectionnés. Une analyse de régression logistique sur chacun des neuf cas ne révèle aucune influence des indicateurs clé du marché du travail. On conclut que le choix du métier est idiosyncrasique; les individus manquent de lucidité dans les choix professionnels et se basent probablement sur des indices en relativement petit nombre. Ces découvertes ont des implications directes sur la pertinence de l’information relative aux professions et sur des questions essentielles qui se posent dans l’exercice de l’orientation professionnelle. Zusammenfassung. Einflussfaktoren bei der Tätigkeitswahl. Diese Forschungsarbeit verwendet einen Wahrnehmungs-Bewertungs-Rückkoppelungsansa tz zur Überprüfung des Verhaltens bei der Wahl von Tätigkeiten unter Bedingungen der Komplexität und der Unsicherheit. Untersucht wird die Behauptung, Die Wahl von Tätigkeiten werde durch sieben implizierte Faktoren bestimmt, nämlich dem Berufsgewicht, dem Anteil der Vollzeit-Beschäftigten, dem Einkommen, den Entwicklungschancen des Arbeitsplatzes, der Geschlechterverteilung in einem Beruf, der berufsspezifischen Arbeitslosenquote und der Altersverteilung der Berufsinhaber. Neun Fallstudien, die das Wahlverhalten bei Präsentation

206

JAMES A. ATHANASOU

von 25 nach Zufallsprinzip ausgewählten Stellenangeboten werden vorgestellt. Die Ergebnisse belegen eine grundsätzliche Eigenwilligkeit im Wahlverhalten. Eine individuelle Analyse zeigte in keinem der 9 Fälle irgendeinen signifikanten Einfluss der genannten Arbeitsmarktdaten auf die Entscheidung. Es wird schlussgefolgert, dass die Tätigkeitswahl individuell verschieden und eigenwillig erfolgt; dass die Einzelpersonen keinerlei Einsicht in das eigene Entscheidungsverhalten haben, und dass ihre Entscheidungen sich wahrscheinlich nur auf einige wenige nicht objektiv belegbare Hinweise stützen. Diese Ergebnisse haben einen direkten Einfluss auf die Bewertung der Bedeutung von Arbeitsmarktinformationen sowie auf zentrale Fragen des Angebots beruflicher Beratung. Resumen. Factores que influyen en la elección laboral. Esta investigación expone un enfoque de la elección laboral basado en lo perceptivo, en la emisión de jucios de valor, y en los refuerzos (Perceptual-Judgemental-Reinforcement approach), bajo condiciones de complejidad e incertidumbre. Se investiga la afirmación de que las elecciones de trabajo se basan en factores implícitos tales como: tamaño específico de la ocupación, la proporción de empleados trabajando a tiempo completo, las retribuciones, las perspectivas laborales, el género predominante en una ocupación, el nivel de desempleo en la ocupación y el grupo de edad predominante en el trabajo. Se presentan nueve estudios de caso que incluyen elecciones de 25 anuncios de trabajos seleccionados al azar. Los resultados indicaron una idiosincrasia substancial en las elecciones de trabajo. Tras el análisis individual de regresión logística realizado, no se encontró ninguna influencia estadísticamente significativa de indicadores clave del mercado laboral en ninguno de los nueve estudios de caso. Se concluyó por tanto que la elección de trabajo era idiosincrática; que los individuos carecían de un conocimiento intutitivo sobre sus elecciones laborales y que probablemente se basaban en algunos indicadores, señales o normas sin explicitar. Estos resultados tienen implicaciones directas relacionadas con la relevancia de la información ocupacional y con aspectos clave de la orientación profesional.

Probably there are myriad views about the factors likely to influence people’s job choices at any point in time. Some of these views might depend on fairly rigid factors such as one’s gender stereotypes of an occupation (e.g., Gottfredson, 1981) or socio-structural influences such as economic or societal power (e.g., Brown, 2000), or one’s personality type (e.g., Holland, 1996) or more generally a fluid system of personal and social influences (e.g., Patton & McMahon, 1999). Whatever might be the mix of factors influencing job choices, some knowledge of how they operate is important for the individual as well as for the community. For instance, Kelly and Lee (2002) investigated career decision problems by analysing the responses of 434 undergraduates to career decision inventories. They described career indecision as “the implicit question” (p. 302) that has direct consequences for the theory and practice of career education, vocational guidance, placement counselling, unemployment services and job informaiton. The purpose of this paper is to investigate job choice in the context of individual judgment. It also extends some new concepts of workplace learning to career development. The paper commences with the idea that decisions about

FACTORS INFLUENCING JOB CHOICE

207

careers hinge clearly upon judgement processes. This research emphasises the individuality of choice but a general framework for conceptualising job choices is also presented and analysed within a social judgement analysis. This paper presents the first application of these principles to job choices.

Career Judgements Any career judgement operates under conditions of uncertainty or complexity. The uncertainty arises from the potential consequences of a job choice for someone’s life. One might anticipate likely scenarios or make plans about the future or even live one day at a time in the hope that this reduces uncertainty. On the other hand, the complexity comes from the huge number of factors one has to take into account (e.g., possible income, potential satisfaction, likely security and stability, probable success, etc.). People may or may not be aware of all these factors or how they analyse them and they may or may not be readily capable always of expressing the balancing or reduction processes – sometimes called judgemental heuristics – that lead to a final decision. Indeed it might be considered astounding that any decisions – let alone satisfactory decisions – are made given the magnitude of the task facing any modern job-seeker or career aspirant or employment seeker. This process of occupational decision making was studied by Mortimer, Zimmer-Gembeck and Holmes (2002) in a series of qualitative interviews as part of a longitudinal study involving respondents surveyed seven years beyond high school and when they were aged around 25 years. They reported that . . . youth referred to becoming ‘increasingly aware’ that an occupation was desirable or that they ‘suddenly realised’ their vocational identity. Such phrasings suggested contextual cues that aid or obscure decisionmaking, but these cues have not been studied . . . many of these cues do not push youth in the ‘right direction’ but are instead ‘eliminators’ that rule out future possibilities . . . and interesting theme for future research is the identification of cues from relationships and organisations that talk to crystallize the decision-making process, especially in contexts marked by uncertainty (2002, p. 462). Accordingly, some means is needed for (a) categorising the likely factors and (b) analysing their potential influence in decisions.

208

JAMES A. ATHANASOU

Figure 1. A Perceptual-Judgemental-Reinforcement model of job choice.

Categorisation How might one categorise the factors that influence choices? Recently, Hager and Halliday (2002) set out a relationship between context, judgement and workplace learning. Their conceptions were translated into a Perceptual-Judgemental-Reinforcement model (Athanasou, 2002) that can also be extended to studying the way judgements are made about jobs (see Figure 1). From a general perspective, it offers a holistic and macro-level framework for analysing behaviour. Briefly, it is proposed that the factors which lead to job choice are either explicit (recognised by everyone) or implicit (often unexpressed, problematic and possibly subliminal). As a starting point one might consider that explicit factors could include the advertised or stated features of a job, such as overtime, shift work, whether one’s own transport was required, stated age requirements such as mature age or junior, award pay conditions, commission, hours of work, whether a job commences immediately, duties (such as stacking, cleaning, delivering fitting or assembly), whether a driver’s licence was required, whether school leavers were preferred or Year 12 completion was needed, whether an experienced employee was required and whether a resume or references were essential. This listing of characteristics was based on a simple content analysis of the advertised features of the 25 jobs used later in this study; no claim is made that the listing of explicit features is exhaustive. Implicit factors might include more subtle labour market factors such as the specific size of the occupation, the proportion of employees working fulltime, the earnings, the job prospects, gender dominance in an occupation, the level of unemployment in the occpation or the predominant age group in the job. While individuals may have some general awareness of these factors, it would be difficult for them to have an accurate detailed knowledge. At

FACTORS INFLUENCING JOB CHOICE

209

the outset it is also recognised that not everyone would agree that these are implicit factors but they are used as a beginning for the analysis of job choice. Of course, choices are also made to to achieve a range of what are called external or internal goods (Halliday & Hager, 2002). External goods might relate to pay, working conditions or other job preferences. Internal goods might relate to deeper values or principles in one’s life, such as honesty, integrity, reliablity, sacrifice, altruism, or just plain ordinary greed, power and influence. The framework is recursive and recognises that some of the expectations might also operate as implicit or explicit factors. The overall representation of the framework divides it neatly into three components but this is a vast oversimplification. Following Hager and Halliday (2002), it is proposed that judgement is central to job choices and that personally relevant factors in a situation as well as purposes affect job choices. One starting point for an investigation of judgement is the relationship between the implicit factors in a situation and job choices. If these factors could be delineated then we can begin to develop a model of decision making for each person. In this study it is proposed to consider how an individual relates to sets of real occupational factors, especially those that might be considered as largely implicit. The study is part of a program of research that seeks to untangle the idiosyncracies of workplace adjustment, individual choice and decision making. The next question is how might one investigate individual judgements about jobs?

Social judgment analysis A methodology for decomposing the contribution of specific factors in judgement has been developed and described but has not been applied to job choices (see Cooksey, 1996). Individuals will differ in their ways of judgement and it is hypothesised this may depend upon the cues or factors in a situation. Figure 2 describes the classic single lens model design for investigating individual judgements as it has been applied in this study. The X1 to X7 in Figure 2 represent items of information that are embedded in an occupational title. As a person reacts to a job they are implicitly responding to these embedded features in the occupation. Some may prefer to call this the stereotype of the occupation. In any event, if we do this enough times then it is possible to correlate the choices with the embedded features or cues in a job or occupation. Multiple regression has been used in social judgement analysis to indicate the overall relationship between the cues and the preferences for occupations and also the weighting of each factor. The individual making the choices would not be aware, however, that they are responding to a series of similarly coded implicit features. The analogy from

210

JAMES A. ATHANASOU

Figure 2. Single lens model for studying judgments.

perceptual research in psychology is that the person is responding to a distal occupation through the lens of a set of implicit proximal cues and that over many trials it really is possible to delineate the relevance and importance of the cues.

Investigating job choice using the lens model In this study it is proposed to randomly select occupations that have been advertised and present these to participants for them to choose which ones (if any) they might be likely to apply. If the occupations are then coded in terms of key features that are implicit or subliminal then it will be possible to describe aspects of each person’s occupational choice. Following Hastie and Dawes (2001, p. 54) it is hypothesised that: 1. Occupational choices rely upon relatively few cues (3 to 5); 2. Judges lack insight into their policies; 3. There are large individual differences in the types of policies adopted. If the thust of the Halliday and Hager (2002) proposal is correct then a person will repond to implicit features of job choices in lawful but idiosyncratic ways. Their idea lends itself to a focus upon intensive investigation of a few individuals. The benefits of intensive case studies of responses to large numbers of real situations lie in the potential for the representative design of an investigation and the ecological validity of the findings. The

FACTORS INFLUENCING JOB CHOICE

211

representative design is one that is authentic (i.e., real jobs) and not artificial as in a questionnaire asking for hypothetical interest preferences. The design has ecological validity because it permits generalisation beyond these circumstances to other job choices made by that person. This idiographic approach (rather than nomothetic) leads to the accumulation of observations and enables us to determine what features of the context are likely to be relevant to the future interests of a person (see examples of this methodology in Athanasou, 1998, 1999; Athanasou & Cooksey, 2001). While most careers research has focused on group studies and preferences, it is the case that the findings can really never be generalised to an individual (unless of course the findings are representative and unanimous). An idiographic perspective, however, takes as its starting point the intensive study of an individual across a range of circumstances. It is assumed that human behaviour might be fuzzy in its parameters but it is essentially lawful. The search is for regularities in a person’s job choice behaviour and from there to extrapolate to other similar circumstances for him/her.

Method Participants The nine participants in this study (N = 7 males; N = 2 females) ranged in age from 15 to 47 years. The median level of their secondary schooling was Year 10 level and one had completed some post-school education to a certificate level. Only one participant was employed. They had undertaken a compulsory vocational guidance assessment to determine their employment potential and they were advised of the purpose at the outset and that involvement was voluntary, anonymous and confidential. The results are reported in a manner that preserves privacy and confidentiality and where necessary any demographic details have been randomly altered to maintain ethical standards of reporting. Instruments Participants were presented with 25 randomly selected (every 50th listing) job advertisements taken from the Federal governments job network (www.jobsearch.gov.au) in the Sydney metropolitan area. The occupations are listed below and a complete description is available from the author upon request.

212 IT sales cement renderer English teacher for Korea gutter/roof cleaner junior trainee salesperson preservation administrator refrigeration mechanic trainee dental assistant

JAMES A. ATHANASOU

store manager window cleaner butcher tiler’s assistant tyre and tube fitter nail technician sales executive senior sales assistant

forklift driver security guard delivery driver dishwasher house clerk junior process worker maintenance fitter panel beater small business traineeship

Procedure The study was part of a one and a half-hour structured vocational assessment comprising mainly an interview dealing with educational, employment social and health factors but also brief standard literacy, cognitive status, psychomotor and interest assessment. Toward the conclusion of the interview, they were shown a randomly selected batch of jobs. Participants were informed that the selection was entirely random and that the source was from the job network’s database. They were asked to indicate which jobs they would find acceptable or would be prepared to select. They were also asked for any reasons that were important for them selecting or not selecting the particular occupations. The occupational choices were coded as 1 and occupational rejections were coded as 0. (A summary of the occupational choices of the nine participants is presented in Appendix A to provide transparency of research and to make available the original data for reanalysis or the findings to be replicated by others, if required.) In addition, each occupation had also been categorised in terms of the following seven characteristics: Occupation size; Earnings (gross); Job prospects; Percentage employed full time; Main age group in the occupation; Percentage who are female in the occupation; and Level of unemployment in the occupation. These details were obtained from the national occupational information database Australian Careers (www.jobsearch.gov.au). All but three of these details were listed exactly as recorded in Australian Careers – the job prospects were recoded (1 = limited; 3 = average; 5 = good; 7 = very good); the lower age limit of the main age group bracket in the occupation was noted (e.g., 15 was used instead of 15–19 years, 25 instead of 25–34 years); and the level of unemployment was coded (1 = low; 3 = below average; 5 = average; 7 = above average; 9 = high). The coded details for the 25 occupations are listed in Appendix A and are also summarised briefly in Table I. Participants were not aware that the occupations had been categorised in this way.

213

FACTORS INFLUENCING JOB CHOICE

Table 1. Descriptive characteristics of the occupations Characteristic

Median

Minimum

Maximum

Occupation size Earnings (gross) Job prospects % employed full-time Main age group % female employees Unemployment level

41900 $601 average 79% 25–34 yrs 20% below average

2400 $481 limited 26% 15–19 yrs 0% low

545800 $1001 very good 98% 55+ yrs 99% high

Analysis The analysis of results was undertaken for each individual. This meant that each person formed their own study comprising a sample of 25 observations or job choices. Therefore, they acted as their own control in terms of social, demographic and personal characteristics. The logistic regression of job choice with the seven implicit factors shown in Table I was calculated. In this analysis, the person’s choices are the dependent variables (binary variables, 1 = chosen and 0 = not chosen) and the seven job characteristics are the independent variables. Multiple regression is used typically in social judgement analyses but cannot be used here because the dependent variable in this study (job choice) is binary and does not meet the underlying assumptions of multiple regression analysis (e.g., normality of errors, homoscedascity, independence). In addition the predicted values from a multiple regression in this case cannot be interpreted meaningfully (e.g., some will be greater than 1 or less than 0). All seven variables were used in the analysis since it was not intended to select variables. Results were interpreted mainly in terms of the significance of the chi-squared values for the regression coefficients. The significant beta coefficients would indicate the cues upon which choices were made and cross-person comparisons would point towards any commonality or idiosyncrasy in the judgement policies adopted. Comparison with the stated reasons would yield some information on any insight into judgement policies. An approximate model R-squared value is reported but it is interpreted with caution. The proportion of correct job choices based on the logistic regression is also reported, using a cut-off value of 0.5.

214

JAMES A. ATHANASOU

Figure 3. Frequency distribution of choices for the nine participants.

Results Overall occupational choices As one might expect, the occupational choices made by each person were few in number. They ranged from 0 to 7, with a median of 2.5. A value of 2.5 selections accounted for around 10% of the occupations listed. The frequency distribution of the number of choices is shown in Figure 3 and this was considered an acceptable range from any randomly selected set of occupations. The following sections consider the results for each person separately. Individual results Person A Person A made six choices that were well above the median number of choices. These included: English teacher for Korea, store manager, trainee dental assistant, house clerk, small business traineeship, and senior sales assistant. Amongst the reasons given for selection were that these jobs did not involve manual work, were “. . . more easier for me to do” and that the person had the requisite “knowledge”. Not one of the regression coefficients for Person A was significant and this was a consistent pattern for all the participants. Nevertheless, the percentage of choices accurately classified by the regression was 100% but it must be remembered that this proportion also includes the jobs that were not selected. Given that most occupations are likely to be rejected then the minimum correct classification in this study could have been 72%, since the maximum number of occupations selected by any one of the nine individuals was 7 out of 25. The adjusted model Rsquared for Person A was .618 (χ 2 (7) = 27.5, p = .0002). These results are also summarised for all the participants in Table 2.

.461∗ 6 92%

.618∗∗∗ 6 100%

Model R-squared # of choices % correctly classified

.330ns 1 100%

–224.424 0.000 0.164 –2.842 0.951 –0.883 0.307 3.659

Person C regression coefficient

Note: the logistic regression model was not calculated for Person G. ∗∗∗ p < .001; ∗∗ p < .01; ∗ p < .05.

22.272 –0.000 –0.013 3.299 –0.373 0.157 –0.269 0.375

–323.824 0.000 0.202 –18.552 1.532 0.491 2.280 –4.084

Intercept Occupation size Earnings Job prospects Percent full time Main age Percent female Unemployment level

Person B regression coefficient

Person A regression coefficient

Variable

Table 2. Log-linear regression coefficients

.248ns 1 96%

–26.402 0.000 0.003 0.809 0.034 –0.204 0.006 –0.119

Person D regression coefficient

.330ns 1 100%

–16.816 0.000 –0.003 –9.752 0.774 –0.372 0.488 –4.775

Person E regression coefficient

.123ns 7 76%

3.119 0.000 –0.000 –0.160 –0.025 –0.048 –0.017 0.208

Person F regression coefficient

.330ns 1 100%

–283.691 0.000 0.437 –54.093 1.522 –2.519 1.552 0.152

Person H regression coefficient

.358ns 4 84%

9.813 0.000 –0.008 2.164 –0.200 0.158 –0.277 –0.641

Person I regression coefficient

FACTORS INFLUENCING JOB CHOICE

215

216

JAMES A. ATHANASOU

Person B Person B made six choices. These included: cement renderer, refrigeration mechanic, window cleaner, dishwasher, tyre and tube fitter and delivery driver. Reasons given for this selection were that these were jobs of which he/she was “capable . . . good, easy jobs, no ‘quals’ (no qualifications required) . . . [not] too old”. The percentage of choices accurately classified by the regression was 92% and the adjusted model R-squared was .461 (χ 2 (7) = 14.5, p = .04). Even though age was mentioned explicitly, it did not enter substantially into the actual choice equation. Person C Person C chose only one occupation, console operator. Reasons for this selection were that “most of them, like lifting stuff; for ladies; know nothing about; junior [age requirement]; no ‘quals’; [no] experience”. The two implicit factors that were mentioned related to gender balance in an occupation and age. The regression coefficients for these factors, however, were not significant. The percentage of choices accurately classified by the regression was 100% and the adjusted model R-squared was .330 (χ 2 (7) = 8.4, p = .29). Person D Person D also made only one choice, that of small business traineeship. The reasons for this selection were: “traineeship . . . learning something new . . . pretty sure able to get into”. No implicit factors related to the occupation were mentioned and the proportion of choices accurately classified was 96%. The adjusted model R-squared was only .248 (χ 2 (7) = 5.6, p = .58). Person E Person E was another participant that made one choice, namely, junior process work (i.e., assembly of circuit boards). Reasons for selection were based mainly on exclusion, namely, “not cut out for office work . . . [occupations rejected] involved bending, lifting”. The proportion of choices accurately classified was 100% but none of the variables in the logistic regression were statistically significant and the adjusted model R-squared was .330 (χ 2 (7) = 8.4, p = .29). Person F Person F made seven choices, which was the largest number recorded. The choices included: IT sales, junior trainee salesperson, preservation administrator, dishwasher, junior process worker, panel beater and small business traineeship. While the choices may appear to be incongruent to an outsider, the stated reasons for selection were that: “things like maybe I would have more of a chance . . . nothing like I really want to do . . .”. The proportion

FACTORS INFLUENCING JOB CHOICE

217

of choices accurately classified was 76% and again none of the variables in the logistic regression were statistically significant and the adjusted model R-squared was only .123 (χ 2 (7) = 2.3, p = .93). Person G In contrast to the other participants, Person G did not make any choices from this sample of occupations and it was unnecessary and impossible to determine the logistic regression in this instance. Person H Person H made only one choice that of sales executive and gave as the reason “don’t think they’re looking for really experienced persons . . . [I] like doing sales things, talking to customers”. The classification accuracy was 84% and again none of the variables in the logistic regression were statistically significant and the adjusted model R-squared was .330 (χ 2 (7) = 8.4, p = .29). Person I The final participant in the group, Person I, made four choices – delivery driver, security guard, forklift driver, senior sales assistant. The stated reasons were that they “. . . like the sound of [the occupation] . . . just want to work thereþ boys’ jobs, wouldn’t want to be a manager”. The classification accuracy was 100% and again none of the variables in the logistic regression were statistically significant and the adjusted model R-squared was .358 (χ 2 (7) = 9.48, p = .21).

Conclusions These nine separate studies showed that there were large individual differences in the types of judgement policies adopted. As expected, participants made idiosyncratic job choices that did not appear to conform to current theories of career development. It was possible to model the judgements with a high degree of accuracy mainly because so many occupational choices are rejected but it was not clear that the participants had any real insight into their judgement policy. Despite the value and importance of labour market factors it did not appear that they affected decisions in any statistically significant way. Yet paradoxically a major thrust of our career services is to provide job relevant information. When occupational choices were justified by qualitative comments they relied upon relatively few cues (3 to 5). Most participants referred to qualifications and training, physical limitations or some gender stereotypes (e.g., boys’ jobs). From the regression coefficients listed in Table II, it was clear

218

JAMES A. ATHANASOU

that job prospects were most important for five out of the nine participants. This leads to a simplistic future hypothesis that job prospects may be a major determinant for guidance of many job choices simply because it may be a realistic, shorthand construct that embodies a number of other job-relevant factors. It was tempting to seek to report demographic and personal details that might somehow correlate with the range and type of choices. However, this would readily identify clients and breach the ethical guidelines. It would be falling into the trap of an ex post facto explanation that is largely untestable. Correlation would not imply explanation. Moreover, even if one knew the causes it is unlikely that one could alter characteristics. It was also tempting to include the advertised features of the jobs but this would mean that the number of cues would have exceeded by far the number of jobs (normally a ratio of 5 or 10 to 1 is recommended for multiple regression and in this study the ratio was already 3.5 to 1). Increasing the number of job choices would have imposed upon the participants and not been possible within the timeframe. At the outset, the Halliday-Hager hypothesis considered that implicit factors influenced judgements. The implicit cues in this study appear to be incomplete and additional factors will need to be proposed if the hypothesis is to remain relevant for job choices. Nevertheless, judgements reflected some use of specific cues or criteria and additional analyses at an individual level may be required. That is, it is recommended that future studies use a standard set of cues (e.g., labour market factors) and then retrospectively include the cues proposed by the person at the conclusion. Another approach may be to explore the use of personal construct theory to elicit the key influences in job choice. The next stage in this program of research has commenced with a larger sample of occupations and inclusion of the explicit features. In summary, these findings established that there is a greater degree of individuality in this process that exceeds the desired uniformity proposed by current theories. Current approaches to providing job information may be predicated on the wrong foundations as it does not appear that individuals share the same cognitive schemas as career practitioners. Hopefully, there is some lawfulness in individual job choice behaviour and decision making. The only way ahead may be to continue a program of intensive individual research that documents the idiosyncrasies of each person’s judgements in different contexts and under varying conditions. This would lead to a useful range of quantitative case studies of job choices and have direct implications for extant theories of career development.

FACTORS INFLUENCING JOB CHOICE

219

Acknowledgement Preparation of this paper was supported by a grant from the Australian Association of Career Counsellors.

References Athanasou, J.A. (2002). The role of contextual factors in judgements: Implications for research into adult learning. Australian Vocational Education Review, 9(2), 1–8. Athanasou, J.A. (1998). Perceptions of interest: A lens model analysis. Australian Psychologist, 33, 223–227. Athanasou, J.A. (1999). Judgements of interest in vocational education subjects. Australian and New Zealand Journal of Vocational Education Research, 7(1), 60–76. Athanasou, J.A., & Cooksey, R.W. (2001). Judgment of factors influencing interest: An Australian study. Journal of Vocational Education and Research, 26(1), 77–96. Brown, M.T. (2000). Blueprint for the assessment of socio-structural influences in career choice and decision making. Journal of Career Assessment, 4, 371–378. Cooksey, R.W. (1995). Judgment analysis: Theory, methods and applications. San Diego: Academic Press. Gottfredson, L.S. (1981). Circumscription and compromise: A developmental theory of occupational aspirations. Journal of Counselling Psychology, 28, 545–579. Hager, P., & Halliday, J. (2002). The importance of context and judgement in learning. In B. Haynes (Ed.), Proceedings of the 30th Conference of the Philosophy of Education Society of Australasia (pp. 15–30). Churchlands: Edith Cowan University. Halliday, J., & Hager, P. (2002). Context, judgement and learning at work. Annual Conference Papers 2002 of the Philosophy of Education Society of Great Britain (pp. 248–256). Oxford: New College. Hastie, R., & Dawes, R.M. (2001). Rational choice in an uncertain world. The psychology of judgment and decision making. Thousand Oaks, CA: Sage. Holland, J.L. (1997). Making vocational choices: A theory of vocational personalities and work environments (3rd ed.). Odessa, FL: Psychological Assessment Resources. Kelly, K.R., & Lee, W-C. (2002). Mapping the domain of career decision problems. Journal of Vocational Behavior, 61, 302–326. Mortimer, J.T., Zimmer-Gembeck, M.J., & Holmes, M. (2002). The process of occupational decision-making: Patterns during the transition to adulthood. Journal of Vocational Behavior, 61, 439–465. Patton, W., & McMahon, M. (1999). Career development and systems theory. A new relationship. Pacific Grove, CA: Brooks/Cole.

220

JAMES A. ATHANASOU

Appendix 1. Job choices Occupation

Person A

Person B

Person C

Person D

Person E

Person F

Person G

Person H

Person I

Butcher Cement renderer Delivery driver Dishwasher English teacher for Korea Forklift driver Gutter/roof cleaner House clerk IT sales Junior process worker Junior trainee salesperson Maintenance fitter Nail technician Panelbeater Preservation administrator Refrigeration mechanic Sales executive Security guard Senior sales assistant Small business traineeship Store manager Tiler’s assistant Trainee dental assistant Tyre and tube fitter Window cleaner

0 0 0 0 1 0 0 1 0 0 0 0 0 0 0 0 0 0 1 1 1 0 1 0 0

0 1 1 1 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 1 1

0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0

0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

0 0 0 1 0 0 0 0 1 1 1 0 0 1 1 0 0 0 0 1 0 0 0 0 0

0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0

0 0 1 0 0 1 0 0 0 0 0 0 0 0 0 0 0 1 1 0 0 0 0 0 0

2. Occupational descriptions Occupation

Occupation size

Earnings (gross)

Job prospects

% full time

Main age group

% Female employees

Unemployment level

Butcher Cement renderer Delivery driver Dishwasher English teacher for Korea Forklift driver Gutter/roof cleaner House clerk IT sales Junior process worker Junior trainee salesperson Maintenance fitter Nail technician Panelbeater

21200 2900 64600 103500 2400 46100 6400 112300 33800 13400 136500 96500 13200 19400

650 951 601 497 1001 641 549 592 945 500 551 760 481 600

3 5 3 3 5 5 1 5 5 1 3 3 7 3

94 87 69 26 49 94 63 66 90 86 59 98 55 98

25 35 35 15 35 35 55 35 25 25 25 35 25 35

1 0 12 59 70 3 20 80 24 19 19 1 94 0

3 3 3 1 5 1 3 3 7 1 1 3 3 5

221

FACTORS INFLUENCING JOB CHOICE Occupation

Occupation size

Earnings (gross)

Job prospects

% full time

Main age group

% Female employees

Unemployment level

Preservation administrator Refrigeration mechanic Sales executive Security guard Senior sales assistant Small business traineeship Store manager Tiler’s assistant Trainee dental assistant Tyre and tube fitter Window cleaner

95600 15300 93200 41900 545800 545800 221800 15300 15700 12600 208800

736 793 720 700 513 513 612 601 521 562 500

7 7 3 5 5 5 3 3 3 5 3

79 95 83 74 34 34 88 91 59 96 36

45 25 25 25 15 15 45 25 20 25 45

69 1 33 17 71 71 40 0 99 4 60

9 7 5 1 5 5 7 5 9 1 1