Mathematics Teachers and Teaching Survey, United States, 2005-2006, and 2015-2016 (ICPSR 39072)

Time Period(s) View help for Time Period(s)

Date of Collection View help for Date of Collection

Data Collection Notes View help for Data Collection Notes

Hide

Study Purpose View help for Study Purpose

The purpose of this study is to address three research questions:

1. To what extent did teachers' MKT change between the 2005 and 2016 samples? Can any observed changes be linked to the teacher characteristics targeted by No Child Left Behind?
2. Has the distribution of teacher MKT become more equitable with regard to student socioeconomic status and race over this time period?
3. Do teachers report greater use of standards-based curriculum materials over this time period?

Sample View help for Sample

In the 2005-2006 survey, SRO decided with University of Michigan School of Education (SOE) researchers to select a nationally representative sample for the mail surveys, using stratified probability proportional to size (PPS) sampling. The target population was all currently operating public schools in the fifty US states and the District of Columbia that had at least ten students in each of the 6th and 7th grades or in each of the 7th and 8th grades.

A total of 21,219 schools from the Common Core of Data (CCD) database met the above requirements and were included in the sampling frame. The frame was stratified by geographic region, according to US Census Bureau regions, and by urbanicity, according to a locale code which is assigned by the US Census Bureau based on geographic location and population attributes. Schools within each stratum were then sorted by zip code prior to sample selection in order to achieve a finer level of geographic stratification.

An initial sample of 1,202 schools was then drawn, in which sample size was allocated to each stratum using a proportionate allocation. Within strata, schools were selected using systematic PPS sampling. The measure of size used was the total number of students in eligible grade levels, as obtained from the CCD. This sample distribution is summarized in Section 5.2.2. For many of these 1,202 schools, QED teacher data provided initial teacher rosters that were confirmed or edited as needed by telephone calls to each school.

After telephone verification of teacher lists was conducted, a final LMT Year 1 stratified sample of 1,000 schools was drawn, from which one teacher per school was randomly selected to be included in the mail survey.

In the 2015-2016 survey, a nationally representative sample of teachers was obtained by first selecting a nationally representative sample of schools. The target population was all currently operating public schools in the fifty US states and the District of Columbia that had at least ten students in each of the 6th and 7th grades or in each of the 7th and 8th grades. A total of 24,270 schools from the CCD database met the above requirements and were included in the sampling frame. The frame was stratified by geographic region, according to US Census Bureau regions, and by urbanicity (for a total of 16 strata). An initial sample of 1,822 schools was then drawn, in which sample size was allocated to each stratum using a proportionate allocation. Within strata, schools were selected using systematic probability proportional to size (PPS) sampling. The measure of size used was the total number of students in eligible grade levels, as obtained from the CCD.

A base roster for each school was built using a combination of data from MCH Strategic Data, a commercial vendor, and individual reviews of school websites. This base roster was then confirmed with each school. Final rosters (i.e. rosters confirmed by the schools) were obtained from 1,584 schools (87%). Rosters were not obtained from schools that refused participation or did not respond. Within the 1,584 rostered schools, teachers were selected at random to participate. The goal was to obtain a completed survey from one teacher per school. If the initial teacher selected refused or did not complete the survey, another teacher within the same school was selected when available. Overall, we obtained 916 completed surveys (57.86% of all rostered schools or 50.2% of all selected schools).

The achieved sample was weighted to account for the probabilities of selection of schools and teachers. This base weight was then adjusted to account for non-response of both schools and teachers. Magnet schools and charter schools were included as part of the target population, but special education schools, vocational schools, and other/alternative settings were excluded. Department of Defense and Bureau of Indian Affairs schools were also excluded as part of the public school target population.

Time Method View help for Time Method

Universe View help for Universe

Middle school mathematics teachers in the United States teaching in the 2015-2016 school year.

Unit(s) of Observation View help for Unit(s) of Observation

Data Type(s) View help for Data Type(s)

Mode of Data Collection View help for Mode of Data Collection

Description of Variables View help for Description of Variables

The Teacher Level Data contains variables about teacher demographics. The Mathematical Knowledge for Teaching Survey - 2005-06 Data and Mathematical Knowledge for Teaching Survey - 2015-16 Data contain variables from the MTK survey administered in 2005-2006 and 2015-2016. The Formal Training in Mathematics Data - 2006 Follow-up Survey contains variables about teachers background information regarding their training. Teacher Curriculum and Material Use Data contains variables about the textbooks that teachers used. The Two Sample Test Data contains data categorizing teachers' prior math education used to compute t tests on teachers MKT score against educational background.

Response Rates View help for Response Rates

Teachers in 58% of rostered schools returned surveys for the 2015-2016 surveys.

Presence of Common Scales View help for Presence of Common Scales

Mathematical knowledge for teaching is measured in a multiple choice format; the answer key documentation can be used to convert these to correct/incorrect answers. Other items (e.g., teachers' use of various resources; teachers' educational background) used Likert or other data collection formats.

Hide

Original Release Date View help for Original Release Date

2025-06-10

Version History View help for Version History

Hide

Weight View help for Weight

A weight variable is included and must be used in order for claims of national representativeness to apply, and to replicate researchers' analyses. Final weights reflect sampling and non-response weights, combined. Please see the user guides for more information.

Hide