SVALUE: Stata module for computing and graphically displaying S-values against their respective P-values
svalue computes S-values corresponding to all P-values ranging from 0.0001 to 1.0, and generates a plot with the specific S-value highlighted. An S-value (Greenland 2018, 2019), also referred to as a self-information or surprisal information measure (Shannon 1948; MacKay 2003; Fraundorf 2019), is a...
| Main Author: | |
|---|---|
| Format: | Software |
| Language: | unknown |
| Subjects: | |
| Online Access: | http://fmwww.bc.edu/repec/bocode/s/svalue.ado http://fmwww.bc.edu/repec/bocode/s/svalue.sthlp |
| Summary: | svalue computes S-values corresponding to all P-values ranging from 0.0001 to 1.0, and generates a plot with the specific S-value highlighted. An S-value (Greenland 2018, 2019), also referred to as a self-information or surprisal information measure (Shannon 1948; MacKay 2003; Fraundorf 2019), is a negative base-2 log transformation of the P-value (or any probability value in the [0,1] range), such that S = -log(p)/log(2). Greenland (2017, 2018, 2019) advocates for the use of S-values to assist in the interpretation of P-values. The S-value is zero (unsurprising) when P = 1.0, increases exponentially as P approaches zero, and can be intuitively yet correctly understood via a simple coin-tossing experiment. As Fraundorf (2019) explains, a bit of surprisal is what you feel after "calling heads" on a coin toss, when the coin lands with heads up! Surprisal is two bits when you throw heads on two of two coins at once. Three bits of surprisal (heads up on three of three coins) is starting to feel respectable. Twenty-four bits of surprisal, on the other hand, is closer to what you experience when winning the lottery. Thus, surprisal reduces the probability of an extremely rare event to a quantity of more manageable size. In the context of P-values, larger S-values correspond to more evidence against the null hypothesis. For example, the corresponding S-value for a P-value of 0.05 is 4.32 bits of information against the null hypothesis (b = 0), which is only slightly more surprising than seeing 4 heads on the first toss of 4 coins (Greenland 2019). Similarly a P-value of 0.0001 has a corresponding S-value of 13.3 bits of information against the null hypothesis (b = 0), which is as surprising as seeing 13 heads in the first toss of 13 coins. Compatibility; Dichotomania; P-values; S-values; Significance; Surprisal |
|---|