Search
Program Calendar
Browse By Day
Browse By Time
Browse By Person
Browse By Unit
Browse By Session Type
Search Tips
Annual Meeting Housing and Travel
Personal Schedule
Sign In
X (Twitter)
In an SCED, a participant is repeatedly observed in both the control and treatment conditions. To establish the internal validity of SCEDs one of the underlying assumptions is the effects of extraneous variables on the dependent variable should be controlled, so that only the independent variable (the treatment) substantially influences the dependent variable. That being said, if the extraneous variables are held constant and the measurement is reliable, there should be little variability of the observations within a phase. In contrast, if the observations within a phase are not stable, it suggests a lack of control over these extraneous factors. In order to help methodologists better understand the features of potential time-varying extraneous variables in SCEDs, this systematic review aims to examine existing empirical SCED studies to determine the prevalence and characteristics of time-varying extraneous variables. Specifically, 50 applied SCED studies were randomly selected and reviewed from the year 2013 to 2017. The raw datasets were extracted and analyzed per Cumming and Schoenfeld’s (1960) criterion to assess the variability and also based on Ferron, Joo, and Levin’s (2017) virtual visual analyst (VVA) algorithm to examine the trends, outliers, and shifts in levels. Results from both visual and statistical analyses elucidate that the majority of the datasets are considered as unstable, suggesting the presence extraneous variables. However, only 13 out of the 50 studies (26%) provide explanations for the instability of their data, and of those, none of them systematically measured the extraneous variables to which they attributed the instability.
Ke Cheng, University of South Florida - Tampa
John M. Ferron, University of South Florida
ZHIYAO YI, University of South Florida