Applying Circular Statistics to the Analysis of Monitoring Data
Abstract
Abstract. Patterns of psychological variables in time have been of interest to research from the beginning. This is particularly true for ambulatory monitoring research, where large (cross-sectional) time-series datasets are often the matter of investigation. Common methods for identifying cyclic variations include spectral analyses of time-series data or time-domain based strategies, which also allow for modeling cyclic components. Though the prerequisites of these sophisticated procedures, such as interval-scaled time-series variables, are seldom met, their usage is common. In contrast to the time-series approach, methods from a different field of statistics, directional or circular statistics, offer another opportunity for the detection of patterns in time, where fewer prerequisites have to be met. These approaches are commonly used in biology or geostatistics. They offer a wide range of analytical strategies to examine “circular data,” i.e., data where period of measurement is rotationally invariant (e.g., directions on the compass or daily hours ranging from 0 to 24, 24 being the same as 0). In psychology, however, circular statistics are hardly known at all. In the present paper, we intend to give a succinct introduction into the rationale of circular statistics and describe how this approach can be used for the detection of patterns in time, contrasting it with time-series analysis. We report data from a monitoring study, where mood and social interactions were assessed for 4 weeks in order to illustrate the use of circular statistics. Both the results of periodogram analyses and circular statistics-based results are reported. Advantages and possible pitfalls of the circular statistics approach are highlighted concluding that ambulatory assessment research can benefit from strategies borrowed from circular statistics.
References
Box, G.E.P. Jenkins, G.M. Reinsel, G.C. (1994). Time series analysis. Forecasting and control . Englewood Cliffs, NJ: Prentice HallChatfield, C. (2004). The analysis of time series . New York: Chapman & HallEbner-Priemer, U.W. Kubiak, T. (2007). Psychological and psychophysiological ambulatory monitoring: A review on hardware and software solutions. European Journal of Psychological Assessment, 23, 214– 226Ebner-Priemer, U.W. Sawitzki, G. (2007). Ambulatory assessment of affective instability in borderline personality disorder: The effect of the sampling frequency. European Journal of Psychological Assessment, 23, 238– 247Fahrenberg, J. Huettner, P. Leonhart, R. (2001). MONITOR: Acquisition of psychological data by a hand-held PC. In J. Fahrenberg & M. Myrtek (Eds.), Progress in ambulatory assessment (pp. 93-112). Seattle: Hogrefe & HuberFisher, N.I. (1993). Statistical analysis of circular data . Cambridge: University PressGottman, J.M. Roy, A.K. (1990). Sequential analysis . Cambridge: University PressHamilton, J.D. (1994). Time series analysis . Princeton, NJ: University PressKrohne, H.W. Egloff, B. Kohlmann, C.W. Tausch, A. (1996). Untersuchungen mit einer deutschen Version der “Positive and Negative Affect Schedule” (PANAS). [Studies with the German version of the PANAS] Diagnostica, 42, 139– 156Kubiak, T. Jonas, C. Weber, H. (submitted). Optimism and perceived social support - An experience sampling study.La Sorte, F.A. Mannan, R. William, R. Richard, T. Grubb, T.G. (2004). Habitat associations of sympatric red-tailed hawks and northern goshawks on the Kaibab plateau. Journal of Wildlife Management Volume, 68, 307– 317Mardia, K.V. (1975). Statistics on directional data. Journal of the Royal Statistical Society Series B, 37, 349– 371Rao, C.R. (1961). A study on large sample test criteria through properties of efficient estimates. Sankhya Series A, 23, 25– 40Reicherts, M. Salamin, V. Maggiori, C. Pauls, K. (2007). The learning affect monitor (LAM): A computer-based system integrating dimensional and discrete assessment of affective states in daily life. European Journal of Psychological Assessment, 23, 268– 277Rocchi, M.B.L. (2001). The maximization of mean vector length for the identification of unknown periods in biological rhythms: An extension to continuous variables. Biological Rhythm Research, 32, 61– 71Rocchi, M.B.L. Ghiandoni, G. (1999). A circular statistics tool for the preliminary identification of unknown periods in biological rhythms: And application. Biological Rhythm Reports, 30, 563– 572Rocchi, M.B.L. Perlini, C. (2002). Is the time of suicide a random choice? A new statistical perspective. Crisis: The Journal of Crisis Intervention and Suicide Prevention, 23, 161– 166Snijders, T.A.B. Bosker, R.J. (1999). Multilevel analysis . London: SageSullivan, H.S. (1953). The interpersonal theory of psychiatry . New York: NortonThompson, L.M. van Manen, F.T. King, T.L. (2005). Geostatistical analysis of allele presence patterns among American black bears in eastern North Carolina. Ursus, 16, 59– 69Warner, R.M. (1998). Spectral analysis of the time-series data . New York: GuilfordWatson, D. (2000). Mood and temperament . New York: GuilfordWatson, D. Clark, L.A. Tellegen, A. (1988). Development and validation of brief measures of positive and negative affect: The PANAS scales. Journal of Personality and Social Psychology, 54, 1063– 1070Wilhelm, P. (2001). A multilevel approach to analyze ambulatory assessment data. In J. Fahrenberg & M. Myrtek (Eds.), Progress in ambulatory assessment (pp. 173-192). Seattle, WA: Hogrefe & HuberWilhelm, P. Schoebi, D. (2007). Assessing mood in daily life: Structural validity, sensitivity to change, and reliability of a short-scale to measure three basic dimensions of mood. European Journal of Psychological Assessment, 23, 258– 267