Furthermore, differences in the probabilities of being sampled as a teacher were accounted for by using teachers' sampling weights (Mplus option 'WEIGHT = TEAsparouhovparouhov 2005).
<.01 <.001 <.001 <.001 <.001 <.001 aAs TIMSS uses a complex sampling design, mathematics teachers' sampling weights (MATWGT) were applied in the descriptive analyses to generate accurate estimates in each participating country.
*As TIMSS uses a complex sampling design, mathematics teachers' sampling weights (MATWGT) were applied in the descriptive analyses to generate accurate estimates on the teacher population in each participating country.
The survey was sent to 387 teachers, sampled from a larger sample drawn from a data tape of nearly 20,000 secondary school biology and life science teachers provided by the National Science Teachers Association in Washington, DC.
Although 88% (open + private acceptance of evolution; choices a + c, Fig. 6) of the New England educators of prospective teachers sampled in this study accepted evolution, their score was below the general faculty (97%, choices a + c, Fig. 6) and comparable to the students' (83%, choices a + c, Fig. 6), the latter higher than their national counterparts (above).
Fig. 3 Latent class analysis of the Czech teacher sample.
Fig. 2 Latent class analysis of the German teacher sample.
Each investigation was coded on several variables describing the child, parent, and teacher samples, as well as reported outcome results.
We removed these teachers from our sample, ending up with 339 teachers in our active teacher sample who participated in random assignment.
An analysis of the German teacher sample, for instance, showed no bias with regard to teachers' gender and their school subjects (Eickelmann et al. 2014a).
