Warren S. Blumenfeld

A Comparison of Achievement Scores of Public-School and Catholic-School Pupils

IN educational circles one frequently encounters questions concerning the relative effectiveness of public-school instructional programs and parochial-school instructional programs. The purpose of this study was to investigate the possibility that there might be group differences in educational achievements for matched groups of public-school pupils and Catholic-school pupils at the eighthgrade level.

Note on the Relationship of Political Preference between Generations within a Household

The relationship of political preference between generations within a household, i.e., parent and child, was quantified by administering a Pmdue Opinion Panel (Gage & Remmers, 1948) questionnaire to 1000 high school students and to one parent of each snidenc, i.e., 1000 parents. T h e s n ~ d e n t sample was representative of the national population of students; the parent sample--defined by the student sample-may be considered representative of the national population of parents of high school s t ~ ~ d e n t s (Blumenfeld, Franklin, & Remmers, 1963 ) . Independent administrations of a political preference item indicated that, of the students, 444 preferred the Democratic party, 312 preferred the Republican party, and 244 were of some other party or undecided; of the parents, 553 reported rhemselves Democrats, 347 were Republicans, and 100 were of some other party or undecided. The student-parent responses were paired for each of the 1000 sets; and the hypothesis of zero interaction between the two item-response distributions was tested by means of chi square. The obtained value was significant beyond the .O1 level (417.77, df = 4 ) ; the coefficient of contingency associated with this test was .54. The significant interaction between the expressed political preferences of students and their parents indicated that close agreement regarding political preference within household was the case. The demonstrated relationship supports the inherent assumpcion as to why the Pzdrdue Opzniolz Panel has been accurate in forecasting national elections through stz~dent responses (Franklin 8: Remmers, 1960) ; and i t corroborates earlier Purdue Opinion Panel findings (Remmers, 1959). I t was concluded chat political preference becween generations within household was positively related.

A Research Note on the Method of Error-Choice

THE method of error-choice has been described by Remmers (1954, p. 239), and it has been applied by Bernberg (1951), Hammond (1948), Kubany (1953), and Wechsler (1950a; 1950b). The method involves providing the subject with an unstructured, somewhat ambiguous, situation (question) in which his choice of item alternatives reveals his attitude. Operationally, this entails providing alternatives equi-distant from the true answer; but not providing a correct answer. The direction of deviation from the non-

PERCEIVED IMPORTANCE AND OBTAINABILITY OF CERTAIN VALUES ASSOCIATED WITH NAVAL SERVICE

A personnel turnover hypothesis involving the congruity of values between the individual and the institution was examined in 2 samples of reenlisted electronics personnel in the Navy. Each sample rated 10 values generally associated with naval service in terms of the importance of the values to them and the obtainability of the values in the Navy. In both samples, the ratings were highly reliable; and certain of the values were rated significantly higher than others. In both samples, the relationship between perceived importance and obtainability was essentially zero. The data neither refuted nor supported the original hypothesis; methodological considerations were discussed.

PERFORMANCE IN TRAINING AND AN ACHIEVEMENT EFFORT RATING

Ss who served in a variety of training experiments were rated by Es in terms of “how hard they tried to achieve on the experimental tasks.” Reliability of the paired comparison procedure was .63. The only strong positive relation with performance, however, was with errors in a single difficult experiment held at the beginning of the program. It is suggested that raters were rating passive “cooperation” instead of “achievement effort.”

Rapidity of Test Completion and Level of Score Attained

Analysis of 4 rapidity indices, an achievement test score, and 2 personality measures indicated that rapidity was a stylistic and reliable response set not significantly related to score level. Personality measures were not related to rapidity, but were slightly related to score level.