Comparison of discontinuous data-collection methods in applied settings

Document Type

Thesis

College

College of Arts and Sciences

Department

Psychology

Degree

M.S. Applied Behavior Analysis

Date Completed

2016

First Committee Member

Ahearn, William

Second Committee Member

Sassi, Jessica

Third Committee Member

Thompson, Rachel

Abstract

"In the current study estimates of the percent occurrence and total duration of behavior were obtained via momentary time sampling (MTS) and partial interval recording (PIR) using sampling intervals of 20 s, 5 min, 10 min, 15 min, and 30 min. These were compared to continuous measures of self-injurious behavior for six participants. Data were collected during the same hour across seven days to examine the accuracy of these time sampling methods in applied settings. The purpose of this study was to examine the feasibility of accurately estimating behavioral occurrence across long time periods (i.e., days, weeks, academic quarters, etc.) using discontinuous measures. The results of the study replicated previous research indicating that PIR overestimated the percent occurrence and duration of behavior and the overestimation increased as the interval increased and that MTS both over and underestimated the percent occurrence and duration of behavior. Estimates obtained with MTS intervals up to 5 minutes somewhat accurately represented the percent of intervals and total duration data in the larger sampling windows and may be a larger time-sampling window to further examine for use in residential settings. However, a great deal of variability was obtained in the accuracy of estimates even with 20-s MTS so caution is warranted when using a discontinuous measure."

This document is available upon request to Western New England University faculty, students, and staff. Please contact D'Amour Library at for access.

Share

COinS