This post should be considered as an addendum to this previous one that discussed causal mediation analysis with a single mediator. That post ended with the argument that causal evidence could be found that procedural justice indeed mediated the effect of the treatment (previous experiences with the police) towards the outcome (normative alignment with the police). This post will discuss two sensitivity analysis techniques which assess this finding’s robustness to unmeasured confounding. Continue reading “ScotCET and sensitivity analysis for causal mediation analysis with a single mediator”
This post finishes the discussion of two other ones (this and this) by providing an example how to carry out causal mediation analysis with a single mediator in R. The “mediation” package is utilised, for a full description of the package’s capabilities, you can refer to Tingley et al. (2014). For STATA users out there, there is a “paramed” package in STATA which should also produce the same results (within rounding error). A note of caution though: the “paramed” package can only be used with linear mediators and outcomes, and with binary logit models with rare outcomes (see: VanderWeele 2016). The dataset for the analysis can be downloaded from the bottom of the page. Continue reading “ScotCET and causal mediation analysis with a single mediator 3. – A demonstration in R”
As discussed in this earlier post, to make meaningful inference from the ScotCET dataset, the focus needs to be shifted to the mediated effect. In such cases, following Baron and Kenny’s (1986) influential article, social scientists usually rely on structural equation modelling and the product method to derive the direct and indirect effects. Nevertheless, this approach has serious limitations that are usually overlooked in the applied literature. Continue reading “ScotCET and causal mediation analysis with a single mediator 2. – Issues with the product method and the sequential ignorability assumption”
This post continues the discussion of the Scottish Community Engagement Trial (ScotCET). In four posts (you can find them here, here, here, and here) I discussed the apparent failure of implementation over the summer. I concluded that despite the unexpected direction of the findings (i.e., those who received the procedurally just treatment had more negative opinion of the police), the treatment effect appeared to be consistent, was not heterogeneous, and thus it could be actually attributed to the research design. Continue reading “ScotCET and causal mediation analysis with a single mediator 1. – Procedural justice and normative alignment with the police”
This is just a quick announcement: with other quant people at Department of Methodology at LSE, we have just started the Quantitative Methods Reading Group. This will be a fortnightly group for PhD students where we can discuss papers on/using advanced quantitative methods from different disciplines in the social sciences. Further details and up-to-date information on the upcoming meetings can be found on the Reading Group’s website.
Should you be interested, please do not hesitate to get in touch, I am happy to add you to the Reading Group’s newsletter.
Being on holiday meant that I have had more time to read the press in the last couple of days. I came across a Guardian/ICM poll which found a shift in public opinion regarding the Brexit bill: compared to April people seem to find a higher EU exit bill more acceptable. Continue reading “How much is too much for the Brexit bill? – Why and when not to read too much into polls”
Teaching the ‘Survey Research Methods: From Design to Analysis’ course at LSE’s Summer School – and my bike being stolen – meant that I’ve been commuting a lot by bus, and had plenty of time to read. In the last couple of days I have been reading “Go Home?”, this short and captivating book. Continue reading “Go Home? – Some thoughts”
This final post finishes the discussion of the previous three (you can find them here, here, and here) by looking at treatment effect heterogeneity and design heterogeneity.
Continue reading “Scotcet and implementation failure 4. – Treatment effect and design heterogeneity with a demonstration in R”
This post continues two earlier ones (this one and this one) and assesses the effect consistency across the matched pairs in the ScotCET dataset. Part of the redacted dataset can be downloaded at the bottom of this page which will allow you to carry out the same analysis presented below. Continue reading “ScotCET and implementation failure 3. – Treatment effect consistency in a blocking design with a demonstration in STATA”
How can one evaluate the veracity of the claim made by MacQueen and Bradford (2017), that the treatment effect emerging from the ScotCET RCT is, in fact, attributable to the experimental design, despite the apparent failure of implementation? For this assessment, one needs to discuss some design features of ScotCET. Continue reading “ScotCET and implementation failure 2. – The design features of ScotCET”