attempt mixed-effects logistic regression. it doesn’t like that.
model1 <- glmer(ChoseShape ~ Condition + TotalChoices + (1|SubjID), combined, family="binomial")
summary(model1)
## Generalized linear mixed model fit by maximum likelihood (Laplace
## Approximation) [glmerMod]
## Family: binomial ( logit )
## Formula: ChoseShape ~ Condition + TotalChoices + (1 | SubjID)
## Data: combined
##
## AIC BIC logLik deviance df.resid
## 250.9 263.8 -121.4 242.9 184
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -0.9589 -0.9520 -0.6351 1.0429 1.5803
##
## Random effects:
## Groups Name Variance Std.Dev.
## SubjID (Intercept) 1.138e-14 1.067e-07
## Number of obs: 188, groups: SubjID, 32
##
## Fixed effects:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.126761 2.147791 -0.059 0.9529
## Condition -0.824078 0.308180 -2.674 0.0075 **
## TotalChoices 0.007123 0.366038 0.019 0.9845
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) Condtn
## Condition -0.001
## TotalChoics -0.995 -0.063
## convergence code: 0
## singular fit
linear regression
model2 <- lm(NumShape ~ Condition + TotalChoices, wide)
summary(model2)
##
## Call:
## lm(formula = NumShape ~ Condition + TotalChoices, data = wide)
##
## Residuals:
## Min 1Q Median 3Q Max
## -1.8674 -0.7599 0.1326 0.4901 2.1326
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 1.1098 2.5199 0.440 0.66289
## Condition -1.1433 0.3801 -3.008 0.00539 **
## TotalChoices 0.2929 0.4311 0.680 0.50219
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.072 on 29 degrees of freedom
## (3 observations deleted due to missingness)
## Multiple R-squared: 0.2422, Adjusted R-squared: 0.1899
## F-statistic: 4.634 on 2 and 29 DF, p-value: 0.01794
regression with poisson distribution because that’s more appropriate for count data
model3 <- glm(NumShape ~ Condition + TotalChoices, wide, family="poisson")
summary(model3)
##
## Call:
## glm(formula = NumShape ~ Condition + TotalChoices, family = "poisson",
## data = wide)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -1.8510 -0.5573 0.0735 0.3167 1.1334
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) 0.2902 1.6258 0.179 0.8583
## Condition -0.5175 0.2437 -2.123 0.0337 *
## TotalChoices 0.1276 0.2770 0.461 0.6450
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for poisson family taken to be 1)
##
## Null deviance: 23.462 on 31 degrees of freedom
## Residual deviance: 18.689 on 29 degrees of freedom
## (3 observations deleted due to missingness)
## AIC: 105.41
##
## Number of Fisher Scoring iterations: 5
