Richard G. Graf

Correlations Redux: Asymptotic Confidence Limits for Partial and Squared Multiple Correlations

Olkin & Finn (1995) developed expressions for confidence intervals for functions of simple, partial, and multiple correlations. This paper describes procedures and computer programs for solving these problems using the methods described by Olkin and Finn. The programs extendthe methods for any number of predictors or for partialing out any number of variables.

Compliant reactions to guilt: Self-esteem or self-punishment.

Summary Two explanations for the consistent finding that compliant behavior increases following experimentally induced guilt were explored. Sixty male and female undergraduates classified as either guilt-induced or control Ss were asked to comply to a request to make phone calls to prospective Ss. In one condition, they were asked to tell prospective Ss a lie, while in the other condition they were merely asked to remind the Ss about the time and place of the experiment. The dependent measure was the number of phone calls an S was willing to make. It was hypothesized that guilty Ss would comply equally to both requests if the compliant behavior was a form of self-punishment, since punishment is a reasonable expectation following a transgression. If compliance were aimed at raising self-esteem, then guilty Ss should comply more to the nondeceptive request. Results indicated that guilt induced Ss were more willing to make nondeceptive phone calls than deceptive phone calls. Controls were equally willing to ...

A New Maximum Likelihood Estimator for the Population Squared Multiple Correlation

Using maximum likelihood estimation as described by R. A. Fisher (1912), a new estimator for the population squared multiple correlation was developed. This estimator ( ρcirc;2 (ML) ) was derived by examining all possible values of the population squared multiple correlation for a given sample size and number of predictors, and finding the one for which the observed sample value had the highest probability of occurring. The new estimator is shown to have greater accuracy than other estimators and to generate values that always fall within the parameter space. The utility of ρcirc; 2 (ML) in terms of providing the basis for the development of small sample significance tests is demonstrated. A Microsoft Excel workbook for computing ρcirc; 2 (ML) and its regions of nonsignificance and for computing a normal transformation for R2 is offered.

Interpersonal Perception as a Function of Help-Seeking: A United States-Netherlands Contrast

This study was designed to test the hypothesis that people have a less favorable attitude toward individuals who seek help from others than toward those who act in a self-reliant manner. Two sets of four narratives were composed in such a way that they differed only as to the sex of the protagonist. Within eact set, the protagonist was faced with one of two different problems (a stalled car or a crying child) which he solved either by himself or by soliciting the help of another person. Dutch and American subjects rated the protagonist of one of the eight narratives on social desirability, general competence, interpersonal competence, psychological adjustment, and physical attractiveness. Among American subjects, protagonists who did not seek help received higher overall ratings. Among the Dutch sample, protagonists received higher ratings for being self-reliant in the stalled car episode, but were rated less highly for not seeking help when dealing with a crying child.

Basic Programs to Determine Sample Size and Exact Power in Anova Designs

Two BASIC programs are described. The first program calculates the sample size per treatment group (n) required to achieve the desired power in any fixed effect ANOVA design. The second program calculates the power to five decimal accuracy when df 1, df 2, alpha, and phi are provided, and can be used to determine the power that will result when an investigator uses some arbitrary sample size.

Behavior in a mixed motive game as a function of induced self-esteem.

High, neutral, and low self-esteem were induced in college students who then took part in a mixed motive game. It was hypothesized that induced low self-esteem would result in highly competitive behavior. This prediction was confirmed for the first trial block of 10 trials but no difference in competitive behavior among the 3 groups was observed during the second through fifth trial blocks. The results were discussed in terms of the success of the induction of level of self-esteem, the perceived strategy of the other player, and the way in which the perceived strategy might interact with chronic and induced levels of self-esteem.

A Basic Program to Compute Fisher's Exact Test

This paper describes a BASIC computer program which performs Fishers exact test for significance of association in 2 by 2 tables. Computations are ordered to avoid overflow, even in samples of one thousand observations or more.