*Replication researchers cite inflated effect sizes as a major cause of replication failure. It turns out this is an inevitable consequence of significance testing. The reason is simple. The p-value you get from a study depends on the observed effect size, with more extreme observed effect sizes giving better p-values; the true effect size plays no role. Significance testing selects studies with good p-values, hence extreme observed effect sizes. This selection bias guarantees that on average, the observed effect size will inflate the true effect size[1]. The overestimate is large, 2-3x, under conditions typical in social science research. Possible solutions are to increase sample size or effect size or abandon significance testing.*

###### [1] By “inflate” I mean increase the absolute value.

###### Figure 1 illustrates the issue using simulated data colored by p-value. The simulation randomly selects true effect sizes, then simulates a two group difference-of-mean study with sample size *n=20* for each true effect size. The effect size statistic is standardized difference, aka *Cohen’s d*, and p-values are from the t-test. The figure shows a scatter plot of true vs. observed effect size with blue and red dots depicting nonsignificant and significant studies. P-values are nonsignifiant (blue) for observed effect sizes between about -0.64 and 0.64 and improve as the observed effect size grows. The transition from blue to red at ± 0.64 is a *critical value* that sharply separates nonsignificant from significant results. This value depends only on *n* and is the least extreme significant effect size for a given *n*.

###### Technical note: The sharpness of the boundary is due to the use of Cohen’s d in conjunction with the t-test. This pairing is mathematically natural because both are *standardized*, meaning both are relative to the sample standard deviation. In fact, Cohen’s d and the t-statistic are essentially the same statistic, related by the identities *d = t**∙sqrt(2/n) *and t* = d**∙sqrt(2/n) *(for my simulation scenario).

###### The average significant effect size depends on both *d* and *n*. I explore this with a simulation that fixes *d* to a few values of interest, sets *n* to a range of values, and simulates many studies for each *d* and *n*.

###### From what I read in the blogosphere, the typical true effect size in social science research is *d=0.3*. Figure 2 shows a histogram of observed effect sizes for *d=0.3* and *n=20*. The significant results are way out on the tails, mostly on the right tail, which means the average will be large. Figure 3 shows the theoretical equivalent of the histogram (the *sampling distribution*) for the same parameters and two further cases: same *d* but larger *n*, and same *n* but larger *d*. Increasing *n* makes the curve sharper and reduces the critical effect size, causing much more of the area to be under the red (significant) part of the curve. Increasing *d* slides the curve over, again putting more of the area under the red. These changes reduce the average significant effect size bringing it closer to the true value.

###### Figure 4 plots the average significant effect size for *d* between 0.3 and 0.7 and *n* ranging from 20 to 200. In computing the average, I only use the right tail, reasoning that investigators usually toss results with the wrong sign whether significant or not, as these contradict the authors’ scientific hypothesis. Let’s look first at *n=20*. For *d=0.3* the average is 0.81, an overestimate of 2.7x. A modest increase in effect size helps a lot. For *d=0.5* (still “medium” in Cohen’s d vernacular), the average is 0.86, an overestimate of 1.7x. For *d=0.7*, it’s 0.93, an overestimate of 1.3x. To reduce the overestimate to a reasonable level, say 1.25x, we need *n=122* for *d=0.3*, but only *n=47* for *d=0.5*, and *n=26* for *d=0.7*.

###### Significance testing is a biased procedure that overestimates effect size. This is common knowledge among statisticians yet seems to be forgotten in the replication literature and is rarely explained to statistics users. I hope this post will give readers a visual understanding of the problem and under what conditions it may be worrisome. Shravan Vasishth offers another good explanation in **his excellent TRN post** and **related paper**.

**his excellent TRN post**

**related paper**

Maybe significance testing and effect size analysis have different levels of importance at different stages in the maturity of a research paradigm. Significance testing is usually about establishing whether there is any sort of effect when one is not sure that it will apply in this case (this population + methodological detail). Once it has been established that an effect can be reliably replicated using a certain procedure and population, attention can turn, through careful empirical work, to establishing “the real” effect size, or more usefully, to exactly how effect size varies for different populations and methodological tweaks.

At that point significance testing is usually superfluous unless there has been a major change in the methodology or population. Just as effect size is pretty superfluous at an earlier point, though it can be helpful to bracket off effect sizes into “large”, “small” and “medium” (which is surely why Cohen introduced these brackets) in order to assess the absolute importance of an exploratory study and the relative importance of its various effects, if several are found.

I would never take the effect size of a single study using a new paradigm as anything more than a vague guess at “the real” effect size, and it is quite surprising to me that anyone would.

And yes I do agree with the central point that these initial effect sizes probably tend to be inflated (though not all of them are, because it’s a stochastic process; thinking for example of the recent mass replications showing that the initial publications of the Knobe effect in moral psychology had underestimated the effect size). I just don’t think that’s a particularly useful point to make at the stage in the research cycle where significance testing is most important.

LikeLike

This a wonderful, nuanced description of the research process. Thanks!

LikeLike