Ideally, high-quality professional development will lead to changes in teaching practice and ultimately to improved student performance. In practice, it will always be difficult to link student outcomes to teacher professional development because of the many determinants of student achievement and the fact that there is no mechanism for examining how teachers are assigned to specific classes. However, the SASS data offer two measures of the impact of teachers' professional development activities. One is teachers' own opinions about the impact of programs and activities in which they have participated. The other is the relationship between teachers' participation in professional development and their reports of their use of various instructional practices.
As described previously, teachers were asked if they had participated in professional development programs on selected topics, including uses of educational technology for instruction, methods of teaching in their subject field, in-depth study in their subject field, student assessment, and cooperative learning. Teachers were then asked whether they strongly agreed, agreed, disagreed, strongly disagreed, or had no opinion on several statements about the impact of those programs. Teachers were reporting on their overall assessment of all professional activities they had participated in on any of these topics, not on each topic separately. Despite widespread criticisms of the state of professional development by researchers and policymakers, in 1993-94 teachers themselves held generally positive views about the impact of at least some of the professional development programs in which they had participated since the end of the last school year.
Overall, 85 percent of teachers who participated in professional development programs on one or more of the above mentioned topics reported that those programs provided them with new information (that is, they agreed or strongly agreed with the statement) (table 19). Sixty-two percent reported that the programs caused them to seek further information or training, and 65 percent reported that they caused them to change their teaching practices. Forty-two percent reported that the programs changed their views on teaching. There were no meaningful differences between public and private school teachers in their assessments (figure 17).
Ten percent of the teachers thought that the programs had wasted their time. Public school teachers were more likely to have this opinion than private school teachers (11 percent versus 7 percent). In the public sector, as years of teaching experience increased, so did the percentage of teachers who thought that the programs had been a waste of their time.
Because of the emphasis on extended duration as a necessary component of effective professional development, it is important to examine whether teachers perceived longer professional development programs as more effective. Therefore, an index of participation was created by multiplying participation (no=0 or yes=1) by the length of the program (8 hours or less=1, 9-32 hours=2, and more than 32 hours=3) and summing across the five types. Thus, an index of 1 would indicate participation in one program for less than a day; an index of 2 would indicate two programs for 1 day or one program for 9-32 hours. Ordinary least squares regression was used to examine the relationship between the level of participation and teachers' assessments of the impact of such participation on their teaching (see appendix C for more details).
The level of teacher participation in professional development programs on the five topics and teachers' assessment about the effectiveness of the programs were positively associated (figure 18). Specifically, the higher the level of participation, the more likely the teachers were to agree or strongly agree that these programs provided them with new information, changed their views on teaching, caused them to change their teaching practices, and made them seek further information or training. This positive association remained significant after taking into account various teacher and school characteristics considered possibly to be related to teachers' assessment. Table B3 (in appendix B) shows the regression results.
Impact on Instructional Practices
Based on teachers' reports in the 1993-94 SASS and the Teacher Follow-up Survey (TFS) conducted the following year, participation in professional development programs appears to be linked to teachers' instructional practices (Henke et al. 1997).27 In particular, participation in professional development was associated with the use of various types of instructional practices that are currently being advocated as effective, such as cooperative learning, portfolios for assessment, and the use of advanced technology in the classroom. Whether teachers participated in the professional development activities and were then motivated to adopt the recommended instructional practices, or whether they sought out professional development activities for help once they decided to implement the practices is unknown, but is not necessarily important. Either link is likely to be valuable to a teacher. It is always possible, of course, that there was no direct connection between the professional development activities and instructional practices, but that teachers who participated in professional development and adopted new teaching methods shared other common traits that prompted them to do both. Nevertheless, the links found between the two are intriguing and worth reviewing.
The TFS asked teachers how often they used various grouping strategies for instructional purposes (whole group, small group, and individual instruction); how often they had students work on group projects (for a group or individual grade); and how often they conducted class discussions of work done by smaller groups. Both public and private school teachers who had participated in professional development programs on cooperative learning in the classroom since the end of the previous school year were more likely than those who had not to use small group instruction at least once a week, assign group projects for individual or group grades, and conduct class discussions of work done in groups (table 20).
Teachers were also asked whether they used portfolios in their classes, and if so, whether or not they included various types of work in them, ranging from traditional assessment tools, such as worksheets, tests, and homework, to tools suitable for evaluating complex learning tasks, such as long-term projects and audio or video work. About half (56 percent) of all teachers used portfolios (table 21). However, teachers who had attended a professional development program on student assessment were more likely than those who had not attended such a program to use portfolios. Again, this was true for both public school teachers (64 percent versus 50 percent) and private school teachers (58 percent versus 46 percent).
Use of Educational Technology
As discussed earlier, teachers are now being called upon to use unfamiliar technologies in the classroom. The TFS asked teachers about their use of various tools in the classroom for demonstrating concepts, including computers, videos, and other electronic media, and about tools that students used in class, such as calculators and computers for writing. Overall, 55 percent of all teachers used computers, videos, or other electronic media; 29 percent used computers for writing; and 24 percent used calculators (which would not be useful in all classes) (table 22). Among both public and private school teachers, those who had participated in professional development on the uses of educational technology for instruction were more likely than those who had not to use each of these tools.
Reading Instruction at the Elementary Level
The link between teachers' professional development and their instructional practices suggested by the SASS and TFS data is supported by data from the 1994 National Assessment of
Educational Progress (NAEP) Reading Teacher Questionnaire (which includes both public and private school teachers). For example, the more staff development hours fourth-grade teachers had in reading, the more likely they were to require students to read from a variety of books (including novels, poetry, and nonfiction) and from materials from other subject areas at least once a month (table 23). Both of these practices are widely encouraged by experts in reading instruction.
Time in staff development in reading was also positively associated with a number of other student activities favored by reading experts, including talking with other students about readings, writing about readings, group activities about reading, discussing interpretations of readings, and explaining understandings of readings (table 24). Moreover, it was negatively associated with workbook exercises, which are generally considered a less productive use of students' time.
Finally, both time in staff development in reading and participation in courses or workshops on assessment in the last 5 years were positively associated with the use of paragraph writings, presentations, and reading portfolios to assess reading progress, and negatively associated with the use of multiple-choice tests (table 25).
Teachers appeared to find the professional development activities in which they had participated since the end of the last school year valuable-at least those on the topics specifically asked about (the use of educational technology, methods of teaching in their field, in-depth study in their field, student assessment, and cooperative learning). Moreover, the greater the intensity of their participation, the more likely they were to think that their professional development experiences had an impact. Finally, there was an association between participation in various types of professional development and the use of certain instructional practices generally linked to contemporary teaching practices or new pedagogical approaches that are thought to be especially effective.
 In the Teacher Follow-up Survey (conducted in 1994-95), teachers were asked about their use of various teaching methods during the past semester. In an analysis first presented in Henke et al. (1997), these data were linked to the professional development activities these teachers reported in 1993-94 that they had participated in since the end of the previous school year. The rest of the discussion of the impact of professional development on instructional practices is based on this previous analysis of the SASS data.