I am testing mediation in SPSS, and all Baron Kenny steps are satisfied, but I have problem in last step. When I include mediator, F becomes non significant, but beta is significant and effect is smaller after including moderator. Can I interpret this like mediation?
Sample size is 126. Dependent variable is disease severity (psoriasis), predictor is Cloninger`s temperament dimension (reward dependence), and mediator is avoidant coping with stress.
Solved – Interpreting mediation, significant beta
mediation
Related Solutions
I have a comment and hopefully an answer.
You use the term "second step", which is a term typically reserved for hierarchical models (e.g., hierarchical regression). Are you certain you are doing mediation analysis? It may help if you describe your question and perhaps how you performed your analyses (e.g., using path analysis, a macro in SPSS). Note that since you have two mediations in your proposed model, it sounds like you would be doing a multi-mediational model, which is something relatively advanced which would likely require path analysis, a programming language, or a macro.
I think the answer to your question is "no", assuming you are doing meditational analysis. For a mediation to be significant, you would need to have a significant direct effect between your mediator and DV, and a significant effect between your IV and mediator, at the very least and in most cases. It doesn't sound like you have this. For a mediation to be significant, your IV needs to "cause" a mediator to such an extent that the mediator's effect on the DV can be at least partially attributed to your IV.
Here is some good basic information on mediation:
The general framework for mediation analysis is to run three models:
- Model 1: a * IV -> DV (to establish some kind of total effect)
- Model 2: b * IV -> MV (to establish some effect on the mediator)
- Model 3: c * IV + d * MV -> DV
where a to d stand for regression coefficients.
The first thing to check for in mediation is evidence of an indirect effect of A on C. The path for that is b (from IV to MV) multiplied by d (from MV to DV with IV in the model). If there is no indirect effect, there is no mediation. If b * d is not statistically significant, then there is not sufficient evidence to detect the mediation effect. I will call b * d, ie for indirect effect: $ie = b \times d$.
Because ie comes from two models, models 2 and 3, we need some method to test whether it is statistically significant. The more traditional approach is the Sobel test, which is just a specialized form of the delta method, which is a first order Taylor Series expansion. The assumption built into this method is the normality of ie. You will often need a large sample size for this assumption to be satisfied. So these days, it is more common to perform bootstrapping. Within each resample, retrieve b and d from Models 2 and 3 respectively, multiply them to get ie, and obtain the confidence interval for ie at the end. Since you use SPSS, there is a macro called Process that implements bootstrapping for the indirect effect.
So if your indirect effect is not statistically significant, you can end your talk of mediation.
If it is statistically significant, it is also reasonable to expect that a in Model 1 would be statistically significant, suggesting evidence of a total effect. The same goes for b suggesting that a at least influences your mediator. Also d suggesting that your mediator influences the outcome. The question then becomes: which type of mediation do you have: partial or complete?
Here's the standard approach:
- If c is statistically significant in the presence of an indirect effect, then the relationship from the IV to the DV persists in the presence of the mediation effect, so we only have partial mediation.
- If c is not statistically significant in the presence of an indirect effect, then the relationship from the IV to the DV vanishes in the presence of the mediation effect, so we have complete mediation.
In your particular situation, I do not see the indirect effect, so it is difficult to begin the conversation about mediation. I will assume that your Steps 3 and 4 are my Model 3. If that is true, then $b=.435$ and $d=.231$, so it is possible to conduct the Sobel test.
With the Sobel test, the formula for the standard error of the indirect effect is: $$\sqrt{b^2 \times se_{d}^2 + d^2 \times se_{b}^2}$$ In your case, $se_d = d/t_d= .231/2.484=0.092995$ and $se_b = b/t_b= .435/4.4=0.09886$. So the standard error of your indirect effect is: $$\sqrt{.435^2 \times 0.092995^2 + .231^2 \times 0.09886^2}=0.04645$$ Your indirect effect will be: $.435\times .231=0.100$. So your $t$-statistic for the indirect effect will be $.1/.04645=2.15$.
I hope this helps. Personally, I doubt the results of mediation analyses. I think it is a causal analysis, so when estimated using ordinary least squares (as you are doing), it is plagued by omitted variable bias. So I do not believe any of these results generally.
Best Answer
I am not an SPSS expert, but from what I am reading from your output you have two models, Model 1 (top) and Model 2 (bottom). Model 2 has an added predictor with a lower F statistic (5.765 --> 2.651). Additionally Model 2 has an R square of 0.067 which is very low. Indicating the model does not have a good impact.
Also, note that for Model 2, your t-statistics are 2.004 (p-val: 0.47) and 1.628 (p-val: 0.106). Both parameters are not significant at the 0.95 confidence level, hence you cannot report
From the reference I've read (http://web.pdx.edu/~newsomj/da2/ho_mediation.pdf)
Your mediation factor (the added predictor) is not significant in Model 2. That said, I would not conclude there is a significant mediation factor. In order for you to conclude mediation, both factors have to be significant. Hence, you have two choices.
Rule out the effect of mediation
Lower your confidence level to 0.90, and then you can conclude mediation.
I'd be very cautious with option 2, as your are now modeling your analysis around trying to obtain significant results, vs trying to be as truthful to the data (and null hypothesis) as possible.
If I've misinterpreted something in the output, let me know and I'll adjust my answer.
Thanks