Sociology, Department of
First Advisor
Kristen Olson
Date of this Version
7-2023
Abstract
While smartphone ownership rates near 90% of the US population, differences in the types of people who do or do not own smartphones may contribute to coverage error biases within smartphone-exclusive web surveys. Therefore, understanding the predictors of smartphone ownership is of interest for survey researchers. Further, differential ownership of smartphone operating systems (OS) such as Android OS and iOS across groups of the population is also of interest for researchers who administer smartphone application surveys which are only compatible with one OS. However, little research has assessed differential rates of smartphone ownership over time nor has any research to my knowledge assessed differential rates of smartphone OS ownership within a general population in the US.
This dissertation addresses three research questions: (1) who owns smartphones in the US and how has smartphone ownership in the US changed over time? (2) How may differential smartphone ownership contribute to coverage error bias in web surveys? (3) Is differential smartphone OS ownership associated with coverage error bias in survey estimates? I use data from the American National Election Studies and the Nebraska Annual Social Indicator Survey to address these research questions. I find that smartphone ownership varies across many characteristics including age, education, and income. While smartphone ownership was found to increase over time, it has not yet reached full coverage of the entire US population. Differential smartphone ownership was found to contribute to coverage error bias in many survey estimates, meaning that a survey of only smartphone owners in the United States would likely yield biased estimates of sociodemographic and sociopolitical outcomes. Finally, I found that Android and iPhone owners differ on only a small number of characteristics such as education and income. Differential ownership of smartphone OS was found to contribute to very little coverage error bias in survey estimates when assuming a target population of smartphone owners in Nebraska. I conclude with implications for survey design and statistical estimation procedures.
Advisor: Kristen Olson
Comments
A DISSERTATION Presented to the Faculty of The Graduate College at the University of Nebraska In Partial Fulfillment of Requirements For the Degree of Doctor of Philosophy, Major: Sociology, Under the Supervision of Professor Kristen Olson. Lincoln, Nebraska: July, 2023
Copyright © 2023 Angelica N. Phillips