Table ES.8. Characteristics of District Stayers, Movers, and Leavers after Two Years by Treatment Status (Percentages Except Where Noted): Two-Year Districts
| Treatment | Control | Difference | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| Teacher characteristic | Stayers | Movers | Leavers | Stayers | Movers | Leavers | Stayers | Movers | Leavers | ||
| College entrance exam scores (SAT combined score or equivalent) | 916 | 1,006 | 1,095 | 967 | 1,040 | 1,081 | -51 | -34 | 14 | ||
| Attended highly selective college | 23.4 | 28.6 | 59.9 | 25.1 | 37.1 | 52.4 | -1.7 | -8.5 | 7.5 | ||
| Major or minor in education | 67.0 | 70.9 | 38.9 | 66.6 | 61.3 | 56.1 | -4.5 | 15.5 | -30.9 | ||
| Student teaching experience (Weeks) | 12.2 | 14.1 | 6.2 | 11.9 | 11.7 | 9.3 | 0.3 | 2.4 | -3.1 | ||
| Entered the profession through traditional four-year program | 61.5 | 76.8 | 25.2 | 66 | 61.3 | 56.1 | -4.5 | 15.5 | -30.9 | ||
| Unweighted Sample Size (Teachers) | 143 | 35 | 25 | 121 | 25 | 15 | |||||
| Unweighted Sample Size (Schools) | 71 | 28 | 20 | 62 | 21 | 13 | |||||
|
SOURCE: MPR calculations using data from the College Board and ACT, Inc.; MPR Teacher Background Survey
administered in 2005-2006, MPR Second Mobility Survey administered in 2007-2008; MPR First and Second
Induction Activities Surveys administered in fall/winter 2005-2006 and spring 2006 to all study teachers. NOTES: Data are weighted to account for the study design. Sample sizes vary due to item nonresponse. The analysis of college entrance exam scores relied on a smaller sample of teachers (143/35/25 treatment stayers/movers/leavers and 121/25/15 control stayers/movers/leavers) and schools (71/28/20 treatment and 62/21/13 control). Stayer: retained in the same school district. Mover: retained in the teaching profession, but not in the same school district. Leaver: no longer teaching. None of the differences between treatment and control stayers, between treatment and control movers, or between treatment and control leavers is statistically different from zero. P-values are suppressed to make the table easier to read. |
|||||||||||