LRR.md
In SingleCaseES: A Calculator for Single-Case Effect Sizes

Parameter definition

The log-response ratio is an effect size index that quantifies the change from phase A to phase B in proportionate terms. It is appropriate for use with outcomes on a ratio scale (i.e., where zero indicates the total absence of the outcome). The LRR parameter is defined as $$ \psi = \ln\left(\mu_B / \mu_A\right), $$ where $\mu_A$ and $\mu_B$ denote the mean levels of phases A and B, respectively, and $\ln()$ is the natural logarithm function. The logarithm is used so that the range of the index is less restricted.

LRR-decreasing and LRR-increasing

There are two variants of the LRR (Pustejovsky, 2018), corresponding to whether therapeutic improvements correspond to negative values of the index (LRR-decreasing or LRRd) or positive values of the index (LRR-increasing or LRRi). For outcomes measured as frequency counts or rates, LRRd and LRRi are identical in magnitude but have opposite sign. However, for outcomes measured as proportions (ranging from 0 to 1) or percentages (ranging from 0% to 100%), LRRd and LRRi will differ in both sign and magnitude because the outcomes are first transformed to be consistent with the selected direction of therapeutic improvement.

Estimation

Denote the sample means from phase A and phase B as $\bar{y}A$ and $\bar{y}_B$ (possibly after transforming to be consistent with the direction of therapeutic improvement), the sample standard deviations from phase A and phase B as $s_A$ and $s_B$, and the number of observations in phase A and phase B as $m$ and $n$, respectively. To account for the possibility that the sample means or sample variances may be equal to zero, even if the mean levels are strictly greater than zero, the LRR is calculated using _truncated sample means and truncated sample variances, given by $$ \tilde{y}_A = \text{max} \left{ \bar{y}_A, \frac{1}{2 D m}\right} \qquad \text{and} \qquad \tilde{y}_B = \text{max} \left{ \bar{y}_B, \frac{1}{2 D n}\right}, $$ and $$ \tilde{s}_A^2 = \text{max} \left{ s_A^2, \frac{1}{D^2 m^3}\right} \qquad \text{and} \qquad \tilde{s}_B^2 = \text{max} \left{ s_B^2, \frac{1}{D^2 n^3}\right}, $$ where $D$ is a constant that depends on the scale and recording procedure used to measure the outcomes (Pustejovsky, 2018).

The LRR is then estimated as $$ R = \ln\left(\tilde{y}_B\right) + \frac{\tilde{s}_B^2}{2 n \tilde{y}_B^2} - \ln\left(\tilde{y}_A\right) - \frac{\tilde{s}_A^2}{2 m \tilde{y}_A^2}. $$ This estimator uses a small-sample correction to reduce bias when the one or both phases include only a small number of observations.

Under the assumption that the outcomes in each phase are mutually independent, an approximate standard error for $R$ is given by $$ SE_R = \sqrt{\frac{\tilde{s}A^2}{m \tilde{y}_A^2} + \frac{\tilde{s}_B^2}{n \tilde{y}_B^2}}. $$ Under the same assumption, an approximate confidence interval for $\psi$ is $$ [R - z{\alpha / 2} \times SE_R,\quad R + z_{\alpha / 2} \times SE_R], $$ where $z_{\alpha / 2}$ is the $1 - \alpha / 2$ critical value from a standard normal distribution.

Percentage change

If desired, LRR estimates and confidence intervals can be translated into the more readily interpretable metric percentage change using the formula $$ \text{% Change} = 100\% \times \left[\exp\left(R\right) - 1 \right]. $$

Primary references

Pustejovsky, J. E. (2015). Measurement-comparable effect sizes for single-case studies of free-operant behavior. Psychological Methods, 20(3), 342–359. https://dx.doi.org/10.1037/met0000019

Pustejovsky, J. E. (2018). Using response ratios for meta-analyzing single-case designs with behavioral outcomes. Journal of School Psychology, 16, 99-112. https://doi.org/10.1016/j.jsp.2018.02.003