Floyd, Marilyn J.

Measurement experts have criticized the use of essay examinations,
chiefly because of their scoring unreliability. Research has emnhasized
scoring inconsistencies and that scorers of essay examinations are frequently
led by composition errors or other writing quality factors to
grade the subject matter--content--of the essays inconsistently. Studies
critical of essay scoring have emphasized the content of essay answers
and have used scorers with minimal expertise in the essays' subject areas.
Recent research reveals that, under certain conditions, i.e., when
writing experts are scoring for placement, using appropriate grading
criteria, reliability rates for essay scoring are improving. Besides
making distinctions between grading conditions and purposes, recent
studies distinguish between content and writing skills, and emphasize
examination of how content characteristics of student essays influence teachers' ratings.
This study focused on whether freshman placement essays would be
graded consistently by different groups of college faculty, experts in
writing and experts in the topic of the essays. Also examined were how
content designated as strong and weak would influence the overall scores
and the scoring of writing quality.

The purpose of this study was to compare the aesthetic tastes of
subjects with formal design training to the aesthetic tastes of the
general population. The Graves Design Judgment Test was used as the
Design Knowledge set of data in this correlational Ex post facto
study. The visual art stimuli were chosen by a panel of art experts
using dual 7 point bipolar scales for realism and design. Visual art
stimuli were reproduced from various worldwide gallery and museum
collections. A diverse set of art stimuli subject matter was used to
minimize content preference by subjects. A multivariate analysis of
variance, a correlation matrix, regression analysis, and principle
components were used to statistically discern a significant difference
between the groups studied. The group with a high cognate knowledge
of design showed a preference for non-objective or abstract art more
than the group without cognate design knowledge. Analysis of variance found a significant difference at the .05 level between the two
groups' preferences for non-objective art but revealed no significant
difference at the .05 level for either group's preferences for
realism. The results of this study are consistent with earlier art
preference research describing differences in trained and untrained
subjects' preferences.

The purpose of this study was to examine the opinions of institutional test administrators, guidance counselors, and students relative to the effects of the College-Level Academic Skills Test (CLAST) on black students in Florida. The attitudes regarding the CLAST test were secured through a thirty-four-statement survey instrument which was distributed to the sixty-seven CLAST testing centers statewide. All sixty-seven institutional test administrators and an equal number of guidance counselors, or some school personnel serving in a guidance position, from these same locations were asked to participate. Two hundred ten students from Florida Atlantic University and its feeder institutions were asked to participate in this study. The feeder institutions selected to participate were: Miami-Dade Community College, Broward Community College, and Palm Beach Junior College. Participants' responses were compared and analyzed. The chi square (X^2) test of significance was used to determine if statistically significant differences in opinions existed among the three groups. Cramer's Phi Coefficient and a modified Pearson's r were used to measure the strength and direction of the differences in opinions. Based on the statistical anlaysis from the data collected, the findings indicated that the status of the respondents had a significant effect on attitudes toward the CLAST program and its fairness to the black student. Institutional test administrators were very supportive of the CLAST program, while guidance counselors were generally nonsupportive. Students tended toward neutrality. This study concluded with this recommendation for further research: a study should be initiated to examine the support programs being provided to the minority students at all educational levels.

The State of Florida mandates the training requirement for the certification of police officers. In recent years, the representation of females and minorities in the training classes of Dade County has increased. Public speculation regarding increased attrition rates and a general lowering of standards, both on the part of the agencies and the training academy, suggested the problem to be studied. The purpose of this study was to investigate the relationship between the scores obtained on the Nelson-Denny Reading Test (vocabulary, reading comprehension and total scores) and the final class average for the 1395 police recruits attending the training classes between 1975 and 1980. Additionally, the study proposed to test differences in the mean reading scores for each year compared with 1975, using sex and ethnicity as control variables, to ascertain whether the scores increased, decreased or remained constant. The design of the study was ex post facto, for which 18 research hypotheses were formulated. Pearson correlational coefficients, a multiple regression analysis, Chi-square analyses and multiple t-tests were performed to test the hypotheses. The analyses of the data indicated there were significant correlations (r = .52, .50 and .55) between the Nelson-Denny test scores and the final class average. The obtained prediction equation yielded only the upper range of final averages, regardless of the reading scores, due to the exclusion of non-completers. Chi-square analyses produced significant results between the high, middle and low reading scores and completion status. The results obtained from the t-test supported the general decline in reading means. Significant differences in reading means were obtained for each level of the independent variables on various pair-wise year comparisons and completion status, but significance was reported more frequently for males, Caucasians and completers in declining scores. The decline in scores was most evident in 1976, 1979 and 1980. In general, the mean reading scores for females were consistently higher than males, and the scores for Caucasians were consistently higher than either Hispanics or Blacks. Additionally, women and Blacks experienced a higher attrition rate.

This study was undertaken because of the increased
role of the building level administrator in the administration
of special education programs at the school level.
' The essential purposes of the study were to assess the
special education competencies of regular elementary,
middle, and high school principals or designees, in a
school-based managed public school system in Florida, and
to locate sources of knowledge/training contributing to
the attainment of competencies. As a basis for the study, eight competencies
previously delineated by Nevin in Vermont in 1977 were
utilized. A portion of the present study sought to
validate the eight competencies as essential for
implementation of special education programs in Florida. Major findings were that most respondents: (1) considered
all eight competencies to be essential, (2) have
not mastered all eight competencies but possessed each to
a degree of understanding, and (3) obtained knowledge/
training most often from on the job experience and wished
to receive further training by in-service provided by the
district.
Competencies not mastered tended to be those
requiring specific knowledge in special education involving
due process, student records, federal and state mandates,
programming policies, and staff functions/qualifications. Several implications relative to competency status,
pre-service and in-service training, and future research
with building level administrators are discussed.