Examining Mode Effects in a Sample of Jewish Households: Differences Beyond Standard Demographics
Submission ID: 5444

Date: Wednesday, 4:30 PM to 6:00 PM 
Session: Session B - W4:30 - 6:00 PM

Primary Presenter

JAZMYNE SUTTON, SSRS

Additional Authors or Round Table Presenters

Eran Ben-Porath, SSRS ,

Cameron McPhee, SSRS ,

, ,

, ,

, ,

, ,

, ,

, ,

Abstract

Studies targeting low incidence populations often require multiple modes of administration in order to effectively reach respondents. In this study, adults in Jewish households in the greater New York area were reached through Address Based Sampling (ABS), stratified by likely Jewish incidence. In order to increase cooperation, a multi-mode multi-contact strategy was employed so that respondents could choose to take the survey online, by phone, or on paper. Because respondents could choose the mode of completion, systematic differences in selection into mode were expected. For example, more than half of phone respondents were age 65 or older compared to less than a third of web respondents. 

However, even when statistically controlling for demographic differences such as age, education, gender, and parental status, significant differences by mode were observed for key outcomes such as feelings of connectedness to the Jewish community, religious participation and membership, experiences with substance abuse, and mental health outcomes.

In this paper, we explore specific factors that may help explain these modal differences. Notably, Jewish denomination significantly contributed to the observed effects by mode. Additionally, issues of social desirability may also be at play when considering differences in response pattern by mode. We consider the implications of these findings for future usage of multi-mode designs in unique low-incidence populations.

Studies targeting low incidence populations often require multiple modes of administration in order to effectively reach respondents. In this study, adults in Jewish households in the greater New York area were reached through Address Based Sampling (ABS), stratified by likely Jewish incidence. In order to increase cooperation, a multi-mode multi-contact strategy was employed so that respondents could choose to take the survey online, by phone, or on paper. Because respondents could choose the mode of completion, systematic differences in selection into mode were expected. For example, more than half of phone respondents were age 65 or older compared to less than a third of web respondents. 

However, even when statistically controlling for demographic differences such as age, education, gender, and parental status, significant differences by mode were observed for key outcomes such as feelings of connectedness to the Jewish community, religious participation and membership, experiences with substance abuse, and mental health outcomes.

In this paper, we explore specific factors that may help explain these modal differences. Notably, Jewish denomination significantly contributed to the observed effects by mode. Additionally, issues of social desirability may also be at play when considering differences in response pattern by mode. We consider the implications of these findings for future usage of multi-mode designs in unique low-incidence populations.