The first order autoregressive model is proposed as a robust model for estimating and testing for means in single subject experiments. It has the advantage of mathematical simplicity, and it provides good approximations to a number of other models of the type typically encountered in behavioral research. Practical considerations on the use of the model are considered including: tests of hypotheses and confidence intervals, sample size requirements, normal approximations, and advantages of the model over the independent error term model. Inferences for means and differences of means are considered.
Get full access to this article
View all access options for this article.
References
1.
BoxG. E. P.TiaoG. C.A change in the level of a nonstationary time series. Biometrika, 1965, 52, 181–192.
2.
ChadwickB. A.and Day, R. C. Systematic reinforcement: Academic performance of underachieving students. Journal of Applied Behavioral Analysis, 1971, 4, 311–319.
3.
EllingstadV. S.WestraD. P.Evaluation methodology for traffic safety programs. Human Factors, 1976, 18, 313–326.
4.
GastwirthJ. L.RubinH.Effect of dependence on the level of some one-sample tests. Journal of the American Statistical Association, 1971, 66, 816–820.
5.
GlassG. V.PeckhamP. D.SandersJ. R.Consequences of failure to meet assumptions underlying the fixed-effects analyses of variance and covariance. Review of Educational Research, 1972, 42, 237–288.
6.
GraybillF. A.Theory and application of the linear model. Belmont, California: Wadsworth, 1976.
7.
JonesR. R.VaughtR. W.ReidJ. B.Time series analysis as a substitute for single-subject analysis of variance design. Paper presented at the annual conference of the American Psychological Association, 1973.
8.
VaughtR. S.JonesR. R.Autocorrelation of time-series data with interrupts. Paper presented at the annual convention of the American Psychological Association, New Orleans, 1974.
