Date of Award

August 2017

Degree Type

Dissertation

Degree Name

Doctor of Philosophy

Department

Information Studies

First Advisor

Iris Xie

Committee Members

Dietmar Wolfram, Laretta Henderson, Susie Lamborn, Dania Bilal

Abstract

This dissertation explores children’s help-seeking behaviors and use of help features when they formulate search queries and evaluate search results in IR systems. This study was conducted with 30 children who were 8 to 10 years old. The study was designed to answer three research questions with two parts in each: 1(a) What are the types of help-seeking situations experienced by children (8-10 years old) when they formulate search queries in a search engine and a kid-friendly web portal?, 1(b) What are the types of help-seeking situations experienced by children (8-10 years old) when they evaluate search results in a search engine and a kid-friendly web portal?, 2(a) What types of help features do children (8-10 years old) use and desire when they formulate search queries in a search engine and a kid-friendly web portal?, 2(b) What types of help features do children (8-10 years old) use and desire when they evaluate search results in a search engine and a kid-friendly web portal?, 3(a) How does children’s (8-10 years old) domain knowledge affect their help seeking and use of help features when they formulate search queries in a search engine and a kid-friendly web portal?, 3(b) How does children’s (8-10 years old) domain knowledge affect their help seeking and use of help features when they evaluate search results in a search engine and a kid-friendly web portal?

This study used multiple data collection methods including performance-based domain knowledge quizzes as direct measurement, domain knowledge self-assessments as indirect measurement, pre-questionnaires, transaction logs, think-aloud protocols, observations, and post-interviews.

Open coding analysis was used to examine children’s help-seeking situations. Children’s cognitive, physical, and emotional types of help-seeking situations when using Google and Kids.gov were identified. To explore help features children use and desire when they formulate search queries and evaluate results in Google and Kids.gov, open coding analysis was conducted. Additional descriptive statistics summarized the frequency of help features children used when they formulated search queries and evaluated results in Google and Kids.gov. Finally, this study investigated the effect of children’s domain knowledge on their help seeking and use of help features in using Google and Kids.gov based on linear regression. The level of children’s self-assessed domain knowledge affects occurrences of their help-seeking situations when they formulated search queries in Google. Similarly, children’s domain knowledge quiz scores showed a statistically significant effect on occurrences of their help-seeking situations when they formulated keywords in Google. In the stage of result evaluations, the level of children’s self-assessed domain knowledge influenced their use of help features in Kids.gov. Furthermore, scores of children’s domain knowledge quiz affected their use of help features when they evaluated search results in Kids.gov. Theoretical and practical implications for reducing children’s cognitive, physical, and emotional help-seeking situations when they formulate search queries and evaluate search results in IR systems were discussed based on the results.

Share

COinS