You can then plot the interaction effect using the following Excel template. can be found on theĭiagonal of the coefficient covariance matrix. Variance of a coefficient is the covariance of thatĬoefficient with itself - i.e. "Covariance matrix" in the "Regression Coefficients" Significant in the regression equation in order forīetween slopes, you should request the coefficient With two-way interactions, the interaction termsĬentred (or standardised) after calculation. Recommended that all the independent variable are
Variables Z and W), run a regression analysis,Īll three independent variables, all three pairs ofĪnd the three-way interaction term. To test for three-way interactions (oftenĪ variable X and dependent variable Y, moderated by And for those readers who use the US version of English, for "centred" and "standardised", read "centered" and "standardized"! (or standardised) variables, you should use theĪnd enter the mean of the variables as zero (and standard deviation as 1 if using standardised versions). Results obtained should be identical whichever Each gives someĪdvantages in interpreting the coefficients - see Dawson (2014)įor more about this (reference below). Some researchers may prefer to use standardised variables. Following Aiken and West (1991), I recommend that for analysis That they have a standard deviation of 1. That are both centred around zero and are scaled so Centred variables are the same as the original version, with the variable mean subtracted so that the new mean is zero. Also worksįor any other outcome using a log link (including negative binomial regression)Ī note about centred and standardised variables. for plotting interactions from generalised for plotting interactions from binary logistic for plotting curvilinear interactions betweenĪ quadratic main effect and a moderator (see can be found on the diagonal of the coefficient covariance matrix.įor non-linear two-way interactions (including generalised linear models), you might want to use one of the following templates:
Note that the variance of a coefficient is the covariance of that coefficient with itself - i.e. If you are using SPSS, this can be done by selecting "Covariance matrix" in the "Regression Coefficients" section of the "Statistics" dialog box. To run simple slope tests, you will also need to request the coefficient covariance matrix as part of the regression output. Therefore where possible, meaningful values should be chosen, rather than just one standard deviation above and below the mean (which is where they will be tested by default if you leave cells B31 & B32 blank) otherwise the tests are arbitrary in nature and quite possibly meaningless. If your moderator is a numerical variable, these are not necessary, but may be useful at specific, theoretically interesting values of the moderator. This template also allows you to perform simple slope tests – these are conditional hypothesis tests of whether the relationship between IV and DV is significant at a particular value of the moderator. If your moderator is binary, ensure you enter the actual values of this variable in cells B31 and B32. If you have control variables in your regression, the values of the dependent variable displayed on the plot will be inaccurate unless you centre (or standardise) all control variables first (although even if you don’t the pattern, and therefore the interpretation, will be correct). You will need to enter the unstandardised regression coefficients (including intercept/constant) and means & standard deviations of the IV and moderator in the cells indicated. Product term should be significant in the regression Recommended that the independent variable and (referred to hence as the IV and moderator) and Variable), first run a regression analysis, You can either use the Excel templatesĭownload them to your computer by right-clicking onĪn independent variable (IV) and dependent variable They use proceduresĪnd Dawson and Richter (2006) to plot theĬase of three way interactions test for significant Templates which help interpret two-way and
0 kommentar(er)
