In this video, we go over the Kurtosis statistic in JASP. The kurtosis statistic will tell us if our data are not normally distributed, or how kurtotic our data is.

A positive statistic will mean a leptokurtic distribution (peaked), a negative one will mean a platykurtic distribution (heavy-tailed), and zero represents no kurtosis.

When it comes to the magnitude of this statistic there are a few different cutoffs or conventions in the literature. Some are more strict and put the cut-off at -1 to 1 (Hair et al., 2017).

Others are less strict and put that cut off at -2 to 2 (Gravetter & Wallnau, 2014) or -1.5 to 1.5 (Tabachnick & Fidell, 2014).

Hair, J. F., Hult, G. T. M., Ringle, C. M., and Sarstedt, M. 2017. A Primer on Partial Least Squares Structural Equation Modeling (PLS-SEM). 2nd Ed. Thousand Oaks, CA: Sage

Gravetter, F. and Wallnau, L. (2014) Essentials of Statistics for the Behavioral Sciences. 8th Edition, Wadsworth, Belmont, CA.

Tabachnick, B. G., & Fidell, L. S. (2014). Using multivariate statistics. 6th Edition, Pearson.

Related Videos:
186: Skewness in JASP [ Ссылка ]
188: Shapiro-Wilk test of normality in JASP [ Ссылка ]

0:00 Intro
0:19 Calculating kurtosis
0:43 Interpreting kurtosis statistica

#JASP #Kurtosis

свернуть