Loyola University Maryland

Emerging Scholars

Yen K. Le, Martin J. Burnham, Takhyun Cho

image divider

Who is Mturk? Personal Characteristics and Sample Consistency of These Online Workers

View the poster >>

Amazon Mechanical Turk (MTurk) is an online data collection service which recruits research participants for small financial payment. MTurk has increasingly attracted the attention of social science researchers as an economical means to gather human research subjects due to its efficiency and low cost in securing data samples. Traditional data collection, relying on undergraduate subject pools, has often confronted researchers with the problem of securing data in a timely and economical fashion. MTurk crowdsourcing initiatives are affording researchers alternatives for collecting data. Previous research has demonstrated that MTurk provides subject pools that move beyond the traditional undergraduate sample to offer wider demographics such as national and international samples of differing age, socioeconomic status, gender and educational levels (Buhrmester, Kwang & Gosling, 2011). Despite increasing utilization of MTurk samples in social science research, spiritual and religious investigations have not been well represented among these studies. The aim of the current study is threefold: 1) to determine the value of MTurk crowdsourcing as a tool for data collection in spiritual and religious inquiry, 2) to validate whether MTurk samples are demographically representative of the general U.S. population, and 3) to measure the repeatability of MTurk samples collected two days apart.

A total of 1003 participants were recruited from Amazon’s MTurk under the restriction that they were U.S. residents with English as their first language and had at least a 95% task approval rating for their previous HITs. Participants were each paid $1.00 U.S. for their participation in the survey and were required to answer all questions in the survey in order to receive their payment. Two samples of participants were gathered over two days (January 5 and 7, 2016); both samples were completed in eight hours. The first sample resulted in 497 valid responses; the second sample yielded 506 valid responses. Participants completed the following measures: International Personality Item Pool 50 (IPIP-50), Assessment of Spirituality and Religious Sentiments (ASPIRES), Affect Balance Scale, Purpose in Life Test, and the Death Anxiety Scale.