interaction between two continuous variables spss

The effect of pred1, then, depends on the value of pred2 and vice versa. Then, repeat the procedure for the second variable. In other words macros may take some initial time to learn but in the long run will save you time. We therefore have little concern about multicollinearity influencing this regression analysis. In terms of statistical efficiency, the popular practice of dichotomising continuous variables at their median is comparable to throwing out a third of the dataset. Figures 9 and 10 present a number of tables of results for both models that are produced by the multiple regression procedure in SPSS. Otherwise I would use values that cover the range (min, mean, max, possibly 2 further values). And my interaction term contains two continuous variables 1) log of employment at the nearest firm 2) log of distance to the nearest firm. Then, use the "Compute" command in SPSS to create a new variable that is the original values minus the mean. The coefficient is statistically significant, based on a p-value of less than .001. This is done in SPSS by selecting the following Menu options: Analyze → Descriptive Statistics → Frequencies. Macros are useful when you need to perform the same statistical procedure for lots of variables or imagine in the future you will be performing the same analysis over and over again. SPSS Inc. was acquired by IBM in October, 2009. How do I report the results of a linear mixed models analysis? ‘Subjects’ indicates the variable containing the case numbers, in this case 1-200 as there were 200 subjects in the study. When conducting a multiple regression with interactions, it is often wise to examine each variable in isolation first. Here both main effects could To my knowledge it is common to seek the most parsimonious model by selecting the model with fewest predictor variables among the AIC ranked models. In the “Simple Scatterplot” dialog box that opens, shown in Figure 12, highlight the new variable, Unstandardized Predicted Values [PRE_1] and click the arrow to move it to the Y Axis text box. The coefficient is statistically significant, based on a p-value of less than .001. We can therefore reject the null hypothesis and conclude that there is evidence of an interaction effect between values of conformity and voting. - Jonas. 5. This example uses three variables from a subset of the European Social Survey (2016): The score variable for immigration was constructed using a summative scale of selected survey items, tested for internal consistency. All rights reserved. Statistical programs, like SPSS, do not always have "point-and-click" commands for every possible statistical test. When I look at the Random Effects table I see the random variable nest has 'Variance = 0.0000; Std Error = 0.0000'. I would use the values mean, mean+SD and mean-SD, only when the SD is somehow sensible. The contour lines are useful to quantify this. In the above example, ‘IQ’ and ‘study’ are the variable names in SPSS given to the IQ and time spent studying by the subjects. Readers are provided links to the example dataset and encouraged to replicate this example. An interaction can occur between independent variables that are categorical or continuous and across multiple independent variables. I have carried out the logodds plot exactly as you said. The analysis revealed 2 dummy variables that has a significant relationship with the DV. Then, insert your variable names into the text, such as. Press OK to produce the graph you have selected. If you did the same, the first set of predicted values will relate to the main effects model. It is the significance of the interaction coefficient, and not the constituent variables, which is of most relevance here. To center the IQ variable, 100 needs to be subtracted from every every subject's IQ score. There is a macro available that will center the variables. We will illustrate the simple slopes process using the hsbdemo dataset that has a statistically significant continuous by continuous interaction. There are three steps involved to calculate the interaction between two continuous variables. We should also produce a frequency distribution of the voter variable. Scores on the immigration attitude scale decrease as scores for priority of conformity values increase. They can all be regarded as continuous, with means of approximately zero and an approximate range of about −2.0 to +2.0 for immigration and −3 to + 3 for conformity. Or in other words, pred1 and pred2 interact with each other. Can anyone provide a good script to plot the interaction of two continuous variables? The advantage of fitting a wiggly surface rather than a plane is that we don’t have to assume that the interaction is linear. Thanks in advance. Survey data was collected weekly. To do this, select the following menu options: Highlight the “Simple scatter” icon shown in Figure 11 and click Define. Thus, there are two continuous variables (X1=IQ, X2=time spent studying), and your dependent variable is the test score (Y=test score). This is called a two-way interaction. These values can no longer be interpreted as the relationship between each independent variable and the dependent variable but are now conditional on the values of the interaction term. However, in this case, the interaction is not explaining a lot more of the variance and a researcher might decide to exclude it in the interests of parsimony. Next, select “Save” from the options on the right-hand side. A significant height x weight interaction would mean that the slope of height on 100m time changes as weight changes: the effect of height might be mitigated at heavier weights. It may be useful to ‘decompose’ the regression surface into the main effects associated with each predictor and the interaction itself. When model fits are ranked according to their AIC values, the model with the lowest AIC value being considered the ‘best’. You can add markers to the graph for those four points by clicking on the graph in the SPSS output window and clicking on the “Add Markers” icon as shown in Figure 13. ‘Subjects’ indicates the variable containing the case numbers, in this case 1-200 as there were 200 subjects in the study. Binning one of the predictors seems appealing since it makes the model easier to interpret. (Note that if you want to compute predicted values from the main effects model as well as the interaction model, you need to select this option and run the analysis for the first model prior to selecting the variables for the second model. A current list of IBM trademarks is available on the Web at “IBM Copyright and trademark information” at http://www.ibm.com/legal/copytrade.shtml. For easy reference, lets called the newly centered IQ score as "IQ_c". (For the purposes of comparison, we saved the predicted values for both models. Describing all of these diagnostic tests is beyond the scope of this example. Interaction effects between continuous variables (Optional) Page 3 Suppose further that 0, 5, and 10 are low, medium and high values of X2. Typically this means your mean score should be entered to say at least 4 decimal places (though the number of decimal places needed will depend on your data). Place your queries over here SPSSZone offers all the SPSS courses from basic introduction to advanced data science in SPSS. The coefficient is statistically significant, based on a p-value of less than .001. This represents the average marginal effect of gender on satisfaction and can be interpreted as the expected change in the dependent variable on average for a one-unit increase in the independent variable, controlling for whether someone voted or not. The edf values in the table above, incidentally, express Is it worthwhile to consider both standardized and unstandardized regression coefficients? This time we want to look at whether levels of positivity towards the benefits of immigration are associated with levels of benevolence and voting in the national election and whether they interact such that the relationship between benevolence and attitudes to immigration differs according to whether you vote or not. We can see this by plotting the results on a graph as shown in Figure 15. If a subject has an IQ of 90, their centered IQ score is -10. However, as I’ll show in this blog post, it’s fairly straightforward to fit and interpret interactions between continuous predictors. This can be done by fitting the main effects and the interaction This opens another dialog box where you can select the plots you want to produce. When estimating a regression model including interactions, we first estimate a main effects multiple regression model. Click Continue to return to the previous dialog box and then click OK to perform the analysis. Moreover, statistical models based on split-up continuous variables are prone to misinterpretation: threshold effects are easily read into the results when, in fact, none exist. Enter the newly centered continuous variable and the categorical variable … In other words, use SPSS, or another statistical program, to find the mean value of the variable. 4. However, in this case, the interaction is not explaining a lot more of the variance, and a researcher might decide to exclude it in the interests of parsimony. In the “Linear Regression:Save” dialog box that opens check “Unstandardized” in the “Predicted Values” section as shown in Figure 5. A -somewhat arbitrary- convention is that an effect is statistically significant if “Sig.” < 0.05. To estimate a regression model with interactions, we need to create a new variable which is the product of the two independent variables. Figure 10 reports a coefficient of .74 for the variable gndr and .82 for voter. As before, the intercept (3.88) can be interpreted as the average level of the dependent variable (satisfaction with the economy) when the values of the independent variables are at zero (in this case for women who did not vote). This page has been accessed 354,957 times. An individual’s mean score across all human values items was subtracted from their mean score on the items relating to the concept in question, thus identifying the priority placed on this particular value in relation to others. It is also useful to explore the possible correlation between your independent variables. Otherwise SPSS will just produce predicted values for the model including an interaction term.). © 2008-2020 ResearchGate GmbH. The negative effect of conformity values on attitudes towards immigration is stronger for those who voted compared with those who did not. Macros are useful when you need to perform the same statistical procedure for lots of variables or imagine in the future you will be performing the same analysis over and over again. In the dialog box that opens, move the voter and gender variables into the “Variable(s)” box and click OK. Screenshots for the procedure to produce histograms in SPSS are available in the How-to Guides for the Dispersion of a Continuous Variables topic that is part of SAGE Research Methods Datasets. As both independent variables are categorical with values of 0 and 1, the model predicts four values, men who voted, women who voted, men who did not vote, and women who did not vote. For this example, just check “Histogram” under the Descriptive heading. software by Daniel S. Soper that performs statistical analysis and graphics for interactions between dichotomous, categorical, and continuous variables. If you have rounded your mean score your centered variable may not have a mean of zero. Middle: pred1 is negatively correlated with the outcome. Other product and service names might be trademarks of IBM or other companies. In a model including an interaction term, the slope estimates cannot be interpreted in the same way, as they are now conditional on other values. In the output, look at the second model in the "Coefficients" box. In this case, we are calling the variable votgnd. This provides estimates for both models and a significance test of the difference between the R-squared values. Left: The predictors are negatively correlated with each other. This is called a two-way interaction. Is it really continuous (all real numbers in [1, 5] are possible) or discrete (values are in {1, 2, 3, 4, 5}) ? An interaction is depicted as a significant value for the interaction variable. We also see that as gender variable increases by one unit (males instead of females), the strength of the slope is indeed reduced with a much gentler slope for males than for females.