Time Series Analysis of Behavioral Intervention

Time Series Analysis of Behavioral Intervention

in Elementary Education Setting

In present days the school administration and teachers are concerned about the growing number of emotional and behavioral problem among students. Hence, they are in favor of adopting an effective behavior management policy to facilitate a positive and productive classroom environment (Baum, 2005). On the basis of several studies, it can be stated that time series methods are most relevant in the psychotherapy field. The reason behind this is the involvement of this method in repeated measurements of individual’s behavior during a baseline phase without any intervention and continuation of the process during intervention and after intervention has stopped. Through this continuous process, time series analysis will be able to provide possible explanation of changes in the client’s behavior as a consequence of intervention (Corduas & Piccolo, 2008).

Extraneous variation or threats to internal validity are emerging due to the consequence of outside events, occurring in our daily life, during the experiment. As the internal validity is identified as the measurement of confidence, manipulated by the independent variable/ treatment/ intervention, therefore extraneous variation is also acknowledged as historical threats to internal validity (Eddy et al., 2003). Possibility may also exist that clients would gain enough maturity during the whole phase to behave positively. Further, a high probability exists, that with the passage of time, the individuals learn to deal with all sorts of emotions and problems for which threat to internal validity enters into maturation phase. Hence, the strength of time series depends on its capability of taking frequent measurements of the dependent variable or the target behavior (Huitema & McKean, 2000). By recording the changes in client’s nature over the time period of baseline phase, this method can be able to accurately predict whether the improvement is due to maturation or not. So, it can be stated that time series research analysis is appropriate for this case study of Kaya. Among the various time series tools, Interrupted Time-Series analysis (ITSA) is considered most significant in determining the effect of experimental manipulation, clinical intervention or even a serendipitous intrusion over temporally ordered scores (Linden, 2015). The technique of Positive Behavioral Interventions and Supports (PBS) has been practiced throughout the experiment to assess the implications of positive strategies for dealing with off-task, physical aggression and verbal aggression behaviors. From the viewpoint of social-practitioner perspective, evidence-based assessments have been done to identify and analyze problems with subsequent implementation of strategies (Wolery, Gast & Hammond, 2010).

On the basis of survey data, conducted during individual activity, group activity and recess period, a graphical presentation can be done to acquire a distinct picture of the impact of services. Serial dependency of time-series model includes stochastic components of time-series observations for the study of the individual behavior. Daily fluctuations around the expected levels of the deterministic effects can be identified with the help of the autocorrelations between observations separated by different time intervals or lags in the series (Barlow, Nock, & Hersen, 2009). In psychological research, predictive models are more preferred for accurate forecasting. So, a dynamic regression model can be obtained by combining the ARIMA approach with regression methods. This hybrid model can be able to subtract the systematic part of the error from each observation (Ramsay et al., 2003). Here, to assess the impact of behavioral intervention, total 24 data are used. Out of which already 3 are given, rest 21 are collected during 7 days on three phases (table attached in appendix). On the basis of the analyses it can be stated that the behavioral intervention strategy has been able to reduce the problems of physical aggression and verbal aggression. Since off-task behavior, with cyclical trend, is not showing significant changes, therefore some more days of counselling is necessary to attain a successful impact in this research study.

Works Cited

Barlow, D. H., Nock, M. K. & Hersen, M. (2009). Single case experimental designs: Strategies for studying behavior for change (3^rd ed.). US: Pearson Education, Inc.

Baum, W. M. (2005). Understanding behaviorism: Behavior, culture and education. Holden, MA: Blackwell.

Corduas, M. & Piccolo, D. (2008). Time series clustering and classification by the autoregressive metric. Computational Statistics & Data Analysis, 52, 1860-1872.

Crosbie, J. (1993). Interrupted time-series analysis with brief single-subject data. Journal of Consulting and Clinical Psychology, 61, 966-74.

Eddy, J. M., Reid, J. B., Stoolmiller, M., & Fetrow, R. A. (2003). Outcomes during middle school for an elementary school-based preventive intervention for conduct problems: Follow up results from a randomized trial. Behavior Therapy, 34, 535-53.

Gottman, J. M. (1981). Time-series analysis: A comprehensive introduction for social scientists. New York: Cambridge University Press.

Huitema, B. E. & McKean, J. W. (2000). Design specification issues in time-series intervention models. Educational and Psychological Measurement, 60, 38-58.

Linden, A. (2015). Conducting interrupted time-series analysis for single- and multiple-group comparisons. The Stata Journal, 15, 480-500.

Ramsay, C. R., Matowe, L., Grilli, R., Grimshaw, J. M., & Thomas, R. E. (2003). Interrupted time series designs in health technology assessment: lessons from two systematic reviews of behavior change strategies. International Journal of Health Technology Assessment in Health Care, 19, 613-23.

Wolery, M., Gast, D. L., & Hammond, D. (2010). Comparative intervention designs. In D. L. Gast (eds.), Single subject research methodology in behavioral sciences (pp. 329-381). New York, NY: Routledge.

Appendix

Here, n is the total number of observations in the series. Z_t is the value of observation at time period i, Z is the mean of series and k is the number of lags.                                     

Table 1: Table on Survey Data during Observation Phases