Table 1. . A linear mixed effects regression (including a by-subject random intercept to account for repeated within-subjects measurements) predicting Log W from Tsimane’ education level and task (computer vs. card version).
| AIC | BIC | logLik | deviance | df.resid |
|---|---|---|---|---|
| 401.8 | 423.7 | −194.9 | 389.8 | 276 |
| Scaled residuals: | ||||
| Min | 1Q | Median | 3Q | Max |
| −2.1824 | −0.6418 | −0.0226 | 0.4943 | 4.9975 |
| Random effects: | ||||
| Groups | Name | Variance | SD | |
| Subject | (Intercept) | 0.02481 | 0.1575 | |
| Residual | 0.20978 | 0.4580 | ||
| Number of obs: 282, groups: subject, 141 | ||||
| Fixed effects: | ||||
| Estimate | SE | t value | ||
| (Intercept) | −1.252289 | 0.041399 | −30.249 | |
| Education | −0.042551 | 0.008400 | −5.066 | |
| task1 | −0.165655 | 0.037229 | −4.450 | |
| Education:task1 | 0.031667 | 0.007554 | 4.192 | |
| Correlation of fixed effects: | ||||
| (Intr) | Eductn | task1 | ||
| Education | −0.681 | |||
| task1 | 0.000 | 0.000 | −0.681 | |
Note: summary(lmer(W_value_lg ∼ Education * task + (1 | subject), REML=F, data=gathered_d)) Linear mixed model fit by maximum likelihood [’lmerMod’] Formula: W_value_lg ∼ Education * task + (1 | subject)