THE EFFECTS OF SELF-SET, PARTICIPATIVELY SET AND ASSIGNED GOALS ON THE PERFORMANCE OF GOVERNMENT EMPLOYEES

A government agency wished to define effective supervisory behavior. Fifty-seven government employees participated in the job analysis. The employees were randomly assigned to one of three goal setting conditions, namely, self-set, participatively set, and assigned goals. The task required each individual to brainstorm individually job behaviors that he or she had seen make the difference between effective and ineffective job behavior as a supervisor. Goals were set in terms of the number of behaviors to be listed within 20 minutes. There was no significant difference in goal difficulty between those with participatively set goals and those with self-set goals. Goal difficulty was held constant between the participative and assigned goal conditions by imposing a goal agreed upon by an employee in the participative condition upon an employee in the assigned condition. There was no significant difference among the three goal setting conditions regarding goal acceptance or actual performance. This was true regardless of employee age, education, position level, years as a supervisor, or time employed in the public sector. The correlation between goal difficulty and performance was .62, .69, and .74, respectively, in the participative, self-set, and assigned goal conditions.     

NUMBER AND SPECIFICITY OF PERFORMANCE OUTCOMES IN THE PREDICTION OF ATTITUDES AND BEHAVIORAL INTENTIONS

The problem of how many performance outcomes to use and how specific they should be in predicting satisfaction and behavioral intentions was addressed. 323 soldiers responded to a desirability and instrumentality scale for each of 16 potential outcomes obtainable from outstanding performance. Scores were factor analysed and composites were formed to reflect each dimension. Three criteria (satisfaction, perceived effort and intention to reenlist) were predicted using (a) all 16 outcome items, (b) only 11 items defining four outcome dimensions and (c) 4 items only, each item reflecting an outcome dimension. In all cases, the 11-item set was a better predictor than the 16-item set, and the 4-item set was nearly as effective as the 16-item set. Instrumentalities were found to be significantly better predictors of satisfaction than of effort, while the reverse was true of valences. It was suggested that adequacy of coverage of the out come domain rather than list length or outcome specificity, was the critical issue in improving predictability.     

Testimony Validity as a Function of Question Form, Atmosphere, and Item Difficulty

The effects of five kinds of questioning, two interpersonal atmospheres of interviewing, and five levels of item difficulty on the accuracy and completeness of testimony about a short film were tested in a legal interrogation setting. Subjects enjoyed the supportive style of interviewing more than the challenging style, but atmosphere had no important effect on recall performance.
The type of questioning produced almost no differences in affective or cognitive reactions. However, as the specificity of questions increased, so did the completeness of testimony. Accuracy of testimony showed slight decreases for more specific questions. The trade-off between accuracy and completeness was mediated by item difficulty. It was very pronounced for items of high difficulty and not apparent for items of low difficulty.
Leading questions by themselves or in interaction with atmosphere did not produce special distortions in accuracy.     

Why Basic Knowledge is Easy Knowledge

The problem of easy knowledge arises for theories that have what I call a "basic knowledge structure". S has basic knowledge of P just in case S knows P prior to knowing that the cognitive source of S's knowing P is reliable.¹ Our knowledge has a basic knowledge structure (BKS) just in case we have basic knowledge and we come to know our faculties are reliable on the basis of our basic knowledge. The problem I raised in "Basic Knowledge and the Problem of Easy Knowledge"² (BKEK) is that once we allow for basic knowledge, we can come to know our faculties are reliable in ways that intuitively are too easy. This raises a serious doubt about whether we had the basic knowledge in the first place.
In "Easy Knowledge", Peter Markie argues that BKS theories do not face any problem concerning easy knowledge.³ I argued that the problem arises in two forms, and Markie takes issue with both. I will argue that Markie's defense of BKS theories fails.