ISPU American Muslim Poll, 2017

Data Archive > U.S. Surveys > General Population > National > Other National Surveys > Summary

Mogahed, D., & Chouhoud, Y. (2020, March 27). ISPU American Muslim Poll, 2017.
Muslims and Jews were surveyed for the Institute for Social Policy and Understanding (ISPU) from Jan. 4, 2017 through Jan. 19, 2017. The study investigated the opinions of Muslims and Jews regarding the 2016 presidential election, the most important issues facing the country, religious discrimination, and domestic violence.
Data File
Cases: 2,389
Variables: 122
Weight Variable: TRIMWGHT
The data from this project are weighted to provide reflections of nationally representative estimates of the adult Jewish and/or Muslim population 18 years of age and older. The weighting process for those from the phone takes into account the disproportionate probabilities of household and respondent selection due to the number of separate telephone landlines and cellphones answered by respondents and their households, as well as the probability associated with the random selection of an individual household member. Specific steps for those collected via phone are provided below:

Probability of Selection (phone number): A phone number's probability of selection depends on the number of phone-numbers selected out of the total sample frame. So for each landline number this is calculated as total landline numbers dialed divided by total numbers in the landline frame and conversely for the cell phone numbers this is calculated as total cell phone numbers divided by total numbers in the cell phone frame.

Probability of Contact: The probability that the sampling unit (households on landlines or respondents on cell phone) will be reached is a product of the number of phones (by type) a respondent or their household answer.

Probability of Respondent selection: In households reached by landline, a single respondent is selected. Thus, the probability of selection within a household is inversely related to the number of adults in the household.

Total Probability of Selection: This is calculated as the phone number's probability of selection (by frame), multiplied by the number of devices of each type the respondent answers, and for landlines, divided by the number of adults in the household. The sample weights derived at this stage are calculated as the inverse of the probability of selection.

The final base-weight is fully calculated for those from the phone portion of this study. Since we are unable to calculate probability of selection for those from the web, those respondents were given a base-weight of 1.

Following application of the above base-weight, the full sample is post-stratified and balanced by key demographics such as age, race, sex, region, education, marital status, number of adults in the household, voter registration, and political party identification within the Jewish and Muslim portions of this study separately. The sample also reflects the distribution of phone usage of the Jewish and/or Muslim population, meaning the proportion of those who are cell phone only, landline only, and mixed users.

Post Stratification Iterative Proportional Fitting ('raking'): With the base-weight applied, the sample will undergo the process of iterative proportional fitting (IPF), in which the sample will be balanced to match estimates of the Jewish and/or Muslim populations determined from three years of data collected through our SSRS Omnibus as well as the PEW estimates. This process of weighting will repeat until the root mean square error for the differences between the sample and the population parameters is 0 or near-zero.

The population parameters used for post-stratification are: age (18-29; 30-49; 50-64; 65+), gender, census region (Northeast, North-Central, South, West), dducation (less than high school, high school graduate, some college, four-year college or more); race/ethnicity (white non-Hispanic or other non-Hispanic; black non-Hispanic; Hispanic); marital status (single; married; other), registered voter (yes/no), political affiliation (Republican; Democrat; Independent/Other), number of adults (one, two, three or more), and phone-usage (cell phone only, landline only, both).
Data Collection
Date Collected: January 4, 2017 to January 19, 2017
Original Survey (Instrument)
ISPU American Muslim Poll, 2017
Funded By
Institute for Social Policy and Understanding
Collection Procedures
CATI interviewers received written materials about the survey instrument and received formal training for this particular project. The written materials were provided prior to the beginning of the field period and included an annotated questionnaire that contained information about the goals of the study as well as detailed explanations as to why questions were being asked, the meaning and pronunciation of key terms, potential obstacles to be overcome in getting good answers to questions, and respondent problems that could be anticipated ahead of time, as well as strategies for addressing the potential problems. Due to the sensitive nature of some of the questions, interviewers were given specific instructions on how to cope with respondents who seemed agitated or distressed by the questions.

Interviewer training was conducted immediately before the survey was fielded. Call center supervisors and interviewers were walked through each question from the questionnaire. Interviewers were given instructions to help them maximize response rates and ensure accurate data collection.
Sampling Procedures
The sampling procedures were designed to efficiently reach the two low-incidence target populations of interest. These are listed below:

1) SSRS pulled sample prescreened as Muslim households from the last four years of its weekly national omnibus survey of 1,000 randomly selected respondents to recontact for this study.

2) SSRS pulled sample prescreened as Jewish households from the last two years of its weekly national omnibus survey to recontact for this study.

3) SSRS purchased listed sample in both landline and cell phone frames from Experian, a sample provider with specific characteristics flagged for each piece of sample. Experian provided sample with flags for Muslim households.

4) Finally, in an effort to supplement the number of Muslim interviews that we were able to complete in the given time frame and with the amount of available prescreened sample, SSRS employed a web panel and completed the final 106 Muslim interviews via an online survey with a sample from a non-probability panel.
Principal Investigators
Dalia Mogahed and Youssef Chouhoud
Bookmark and Share