John R. Slate

Mixed data analysis: Advanced integration techniques

Abstract The purpose of this paper is to provide a coherent and inclusive framework for conducting mixed analyses. First, we present a two-dimensional representation for classifying and organizing both qualitative and quantitative analyses. This representation involves reframing qualitative and quantitative analyses as either variable-oriented or case-oriented analyses, yielding a 2 (qualitative analysis phase vs. quantitative analysis phase) × 2 (variable-oriented analysis vs. case-oriented analysis) mixed analysis grid. We present a comprehensive list of specific qualitative (e.g. method of constant comparison) and quantitative (e.g. multiple regression) analyses that fit under each of the four cells. Next, we provide an even more comprehensive framework that incorporates a time dimension (i.e. process/experience-oriented analyses), yielding a 2 (qualitative analysis phase vs. quantitative analysis phase) × 2 (particularistic vs. universalistic; variable-oriented analysis) × 2 (intrinsic case vs. instrumental case; case-oriented analysis) × 2 (cross-sectional vs. longitudinal; process/experience-oriented analysis) model. Examples from published studies are presented for each of these two representations. We contend that these two representations can help mixed researchers – both novice and experienced researchers alike – not only classify qualitative, quantitative and mixed research, but, more importantly, can help them both design their mixed analyses, as well as analyze their data coherently and make meta-inferences that have interpretive consistency.

Beginning teacher success: an investigation into the feedback from mentors of formal mentoring programs

Teacher mentors of first‐year teachers provided insight into those practices they viewed as essential for their success in the mentoring role. Specifically, they were queried about teacher involvement/support, staff development, administrative support and resource materials. Almost all of the mentor teachers believed a teacher mentoring program that had well‐defined goals was necessary for retaining beginning teachers. Some of the mentor teachers considered that staff development that provided strategies to serve students in special populations better was important for retaining beginning teachers. Mentors indicated that the most difficult parts of their duties involved scheduling conflicts with the mentee, receiving little support from administration (e.g., limited release time to meet with beginning teachers), and having no guidelines or preparation for what they were expected to do. Additional comments made by mentors consisted of the need for more time for new teachers to be able to reflect upon their teaching practices and a de‐emphasis on Texas Assessment of Knowledge and Skills (TAKS) scores. They also indicated that because too much time was being spent on students’ standardized test scores, schools and schooling were being negatively influenced.

Conducting mixed analyses: A general typology

Abstract In this article, we provide a typology of mixed analysis techniques, namely the Mixed Analysis Matrix, that helps researchers select a data analysis technique given the number of (a) data types collected (i.e. quantitative or qualitative; or quantitative and qualitative) and (b) analysis types used (i.e. quantitative or qualitative; or quantitative and qualitative)—yielding a 2 X 2 representation involving four cells that each contain specific analytical techniques, with two of these cells containing a total of 15 mixed analysis techniques. Furthermore, we describe the fundamental principle of mixed analysis, describe the steps in a mixed analysis, and delineate the rationale and purpose for conducting mixed analyses. For each technique, readers are directed to published studies that serve as illustrative examples. Outlining the mixed-analysis techniques available for researchers hopefully will increase awareness of the number of choices for analyzing data from mixed studies.

What Makes a Good Elementary School? A Critical Examination

Abstract With a renewed emphasis on student achievement, school improvement, and shared decision making, and in an effort to create and maintain effective schools, researchers try to delineate those characteristics that positively affect student learning and behavior. In exploring the different views that persons have regarding the characteristics of effective schools, the authors reviewed research studies to identify factors considered to be determinants of an effective school. Specific topic areas examined were the views of teachers, parents, and administrators regarding effective schools, as well as an examination of the related research literature. Consistent across the 4 areas was the importance of a positive school climate and strong leadership. Implications for educational leaders are provided.

The "Digital Divide": Hispanic college students' views of educational uses of the Internet

Concern has been expressed recently regarding the presence of a digital divide between majority and minority groups. Surveyed were 226 Hispanic college freshmen enrolled at a doctoral intensive institution in the Southwest regarding their attitudes toward educational uses of the Internet and their uses of technology. Statistically significant differences were found between males and females in their attitudes toward and uses of the Internet and computer technology. Differences were also revealed between students whose primary language spoken at home was English and students whose primary language spoken at home was Spanish in their attitudes toward and uses of the Internet and computer technology. Interestingly, no differences were present between first-generation and non-first-generation college students. Implications of our findings and relationships to existing literature are provided.

Transracial adoption of Koreans: A preliminary study of adjustment

Because of the increased incidence of transracial adoption, we explored psychological issues of self-concept and acculturation for a sample of 175 Korean adoptees who completed the following instruments: Self-Description Questionnaire-III, Suinn-Lew Asian Self-Identity Acculturation Scale, and the Asian American Cultural Identity Seale. Significant differences were found between our sample and the normative sample on the SDQ-III, with our sample scoring higher in Religion, Honesty, Relations with Opposite Sex, Physical Appearance, and General self-concept and lower in Math, Emotional Stability, and Relations with Parents. The later the age at which Korean adoptees were placed, the higher their Math and Honesty self-concept subscores were. The Verbal self-concept score, however, was negatively related with age of placement of Korean adoptees. In general, our findings suggested that certain aspects of self-concept were influenced by early or later placement of Korean adoptees. Implications of findings are discussed.

Who's Taking the Advanced Placement Courses and How Are They Doing: A Statewide Two-Year Study

Enrollment in Advanced Placement courses in all high school campuses in the State of Texas was examined for the 2004-2005 and the 2005 2006 school years to determine the extent to which student diversity was present. A higher percentage of the female student body, 17%, was enrolled in Advanced Placement courses than was the male student body, 13%. Almost a fifth of the White student body was enrolled in Advanced Placement courses, compared with slightly over 10% of the Hispanic student body and the African American student body. The percentages of economically disadvantaged students, Limited English Proficient, and students with special learning needs were low, as anticipated. Student success in these courses, however, differed as a function of gender and ethnicity, with African American students experiencing the least success. Of interest was that Hispanic students demonstrated the highest degree of success in these courses. Implications of these findings are discussed.

High School Students and Their Lack of Preparedness for College: A Statewide Study

In this study, the authors examined the college-ready graduate rates of all students (n = 1,099 high schools) in the State of Texas for the 2006-2007 school year. Data were analyzed for students’ scores in reading, in math, and in both subject areas combined. Approximately one-third of all students were determined to be college-ready in both subject areas. Statistically significant and practically relevant differences, reflecting moderate to large effect sizes, were present in reading, math, and both subjects among Hispanic, African American, and White students. Concerns are expressed about the lack of preparedness of students for college and about the presence of strong achievement differences as a function of ethnicity. Implications of these findings are discussed.

Practitioners' administration and scoring of the WISC-R: Evidence that we do err

Abstract WISC-R protocols (n = 56) completed by one certified and eight licensed practitioners were analyzed to determine whether and to what degree administration and scoring mistakes occured. Numerous mistakes were observed, such as failure to record examinee responses, assigning too few or too many points to answers, inappropriate questioning, and failure to obtain correct basals or ceilings. Subtests and subtest items on which practitioners are prone to err were identified. Examiner errors, when corrected, resulted in changes in the Full Scale IQ that had the potential to influence labeling and placement decisions. The implications of these errors for professional practice are discussed and suggestions for minimizing examiner errors through better training are provided.

Examiner Errors in Administering and Scoring the WPPSI-R

We investigated administration and scoring errors on 57 WPPSI-R protocols completed by 7 examiners. These examiners made frequent errors, such as failing to record examine responses, assigning incorrect point values to examinee responses, and determining incorrect basals and/or ceilings. When examiner errors were corrected, Full Scale IQs were changed on 53% of protocols. The incorrect IQs assigned by examiners resulted in one potential diagnostic error. Examiner errors appeared to stem from three sources: examiner carelessness, inadequate training, and complex scoring procedures.