Student And Recruiter Insights On The Importance Of Job Attributes [*].

The ability to create a strong person-organization fit (Adkins et al., 1994; Cable and Judge, 1994; Judge and Bretz, 1992) depends to a considerable extent on good communication and self-insight. Studies in the accounting literature (Bundy and Norris, 1992; Carpenter and Strawser, 1970; Kirsch et al., 1993) have sought to provide evidence on this issue by asking (i.e., surveying) accounting students about the importance of various job attributes. Unfortunately, people often have very little insight into how important or unimportant various factors are in their decision process. For example, it is well known that people, when asked to assign weights to decision variables, will underweight the most important ones and overweight the least important variables (Slovic and Lichtenstein, 1971). This lack of understanding about what is or is not important has a direct bearing on a student's ability to report what makes a job attractive. As Judge and Bretz point out, priming artifacts, "... such as the tendency to ra te pay as less important than it really might be" (1992: 262), often result from these direct estimation designs. Survey results of senior accounting students may show particular variables as important, but careful, systematic elicitation of a variable's importance may reveal something completely different.

Students are likely to accept a job offer based on whether the variables that they perceive to be important are present in that job opportunity. One responsibility of the recruiter is to communicate the attributes of a particular job clearly. To the extent that the recruiter "knows" what is really important to the student, this information about the job can be provided. Further, the recruiter could educate the student about other attributes of the position that the student may not otherwise consider. Lack of insight by the student and/or the recruiter about what is really important may lead to poor job choices, job dissatisfaction and eventual turnover. The effects of job dissatisfaction and excessive turnover are clearly factors that a firm would like to avoid if possible.

RESEARCH QUESTIONS

Surveys, as a method of eliciting information from students about their job selection preferences, have been used often throughout the years (e.g., Bundy and Norris, 1992; Carpenter and Strawser, 1970; Shivaswamy and Hanks, 1986; Yunker et al., 1986). These studies have yielded results that have important implications for accounting practice. For example, one conclusion that could be drawn from this line of research is that accounting students generally do not regard salary as the most important factor in choosing their first jobs and that factors such as job security, interesting work and opportunities for advancement are more important (Bundy and Norris, 1992). (An extensive review of the literature on recruiting may be found in Rebele et al. (1991) and in Rebele et al. (1998).)

The primary advantage of survey research is a "... wide scope: a great deal of information can be obtained from a large population" (Kerlinger, 1986: 387). Weaknesses in this methodology come from the fact that it does not go much below the surface, and it can be time consuming. We believe that a third weakness is more likely to make the results of surveys suspect; namely, this method "... lifts the respondent out of his own social context ...." (Kerlinger, 1986: 387). Answers to the questions (or ratings of the attributes) may be what the student thinks the researcher wants to hear or what the student has been told should be important.

Asking students to describe what is important to them requires that they understand the cognitive processes behind their judgments, evaluations and choices. Nisbett and Wilson (1977) reviewed the evidence on introspection of this kind and found almost no evidence to support people's ability to access higher-order cognitive processes. What may be happening is that "... their reports are based on a priori, implicit causal theories, or judgments about the extent to which a particular stimulus is a plausible cause of a given response" (Nisbett and Wilson, 1977: 231). In other words, people will describe the process as they believe it should be (in theory).

An alternative to asking participants to reveal their cognitive processes through surveys is to measure their cognitive processes through judgment modeling. This approach involves providing participants with a given set of cues (here, job attributes) for their use in making a set of decisions (in this case, job evaluations). The effect (i.e., weight) of each cue on the participant's evaluations can then be determined statistically. (See Slovic and Lichtenstein (1971) for a description of judgment modeling using the analysis of variance paradigm.) Comparing the attribute weights obtained through judgment modeling to the participant's assigned weights is a study of self-insight. Slavic and Lichtenstein reviewed the literature on self-insight and concluded that there were "... serious discrepancies between subjective and computed relative weights" (1971: 683).

In business contexts, early studies of self-insight examined stockbrokers (Slovic, 1969; Slovic, 1972; Slovic et al., 1972) and auditors (Ashton, 1974; Joyce, 1976). Trained auditors would know based on "... a priori, implicit causal theories" (Nisbett and Wilson, 1977: 231) acquired from reading a book that separation of duties is important. So, even in situations where self-insight looks to be significant, introspection in fact may not have occurred. The known rules may have simply been applied.

Students

The job evaluation and selection decision is one in which there are no formal rules. A student may only be advised that the "quality of the organization" is important, or that "the money will come." Here, we would not expect students to exhibit a high degree of self-insight. The research question of whether students have self-insight about how they evaluate jobs may be stated as a null hypothesis:

[H.sub.1]: There is no difference in the average statistical weights and self-insight weights for the six job attributes that define each job opportunity.

The alternative hypothesis is that the average statistical weights will differ from the self-insight weights, indicating a lack of self-insight.

Recruiters

Recruiters, based on their experiences, should have knowledge of the factors that are truly important when considering a job opportunity. When a recruiter and a student enter into an interviewing situation, however, the normative weights that should be assigned to the job attributes are not at issue. The student has already internalized what he or she believes to be important in the potential job.

Two elements of a recruiter's job are those of a quality filter and as an advocate for the firm. As a quality filter, the role of the recruiter is to get the best students from the available pool. As an advocate for the firm, the recruiter must convince these students that his or her firm would provide a good "fit." In both cases, knowledge of what the student wants or thinks he or she wants is essential. A recruiter can "sell" the firm if it is strong in what the student sees as important, or can attempt to "educate" the student if the recruiter believes that the student's priorities are not as they should be.

A recruiter with experience, having identified systematic influences, may have more insight into how students make job choices than does the student. As a group, recruiters, assessing how much weight they think the job-seeking students they typically recruit would place on each of the job attributes, may come very close to either the statistical weights or the students' self-insight weights. If the recruiter wants to make a true and long-lasting match for the firm, the recruiter will focus on the underlying weights and try to educate the student on what is necessary for a good fit. In this case, recruiters should have a well-honed sense of the student's statistical weights. However, if the knowledge of the student's perception of the important attributes is decisive (i.e., the recruiter wants to "sell" the firm), then the recruiter will understand and focus on the perceived (self-insight) weights of the student.

To address how well the recruiters know the students, we will test the following null hypotheses:

[H.sub.2]: There is no difference in the average statistical weights and recruiter assessed weights for the six job attributes that define each job opportunity.

[H.sub.3]: There is no difference in the average self-insight weights and recruiter assessed weights for the six job attributes that define each job opportunity.

Alternative hypotheses are that there is a difference between the respective weights, indicating a lack of insight by the recruiters into the students' actual or perceived weights.

METHOD

We describe in this section the participants, the process used to select the variables used in the task, the collection of data, and the method of analysis. Our aim is to find students who are highly motivated to understand their decision-making process, and recruiters who are very familiar with this group of students. Six job attributes that are perceived as falling along a wide range of a continuum of importance in the job decision were used to construct hypothetical job opportunities. A method for calculating statistical weights and self-insight assessments is then described.

Participants

There are many levels of accounting students and recruiters who recruit them. To achieve an efficient cross-section of all accounting students, we would need to match individual students across levels with the recruiters most likely to interview them. For the sake of parsimony, we selected one well-defined group of accounting students, Beta Alpha Psi members (an honorary accounting fraternity), and recruiters from firms that specifically target these members or students of equal quality. Accounting students who are members of Beta Alpha Psi must maintain a grade-point average of 3.0 (where A is equal to 4.0). Further, because of the participation requirements inherent in being a member, the students must be highly motivated to achieve success.