First, activity in the two regions of interest (the supplementary motor area and the parahippocampal gyrus) identified during the localizer scanning was quantitatively characterized (with the use of the average
generalized linear model estimate for each region of interest).
Not exact matches
We built a
generalized estimating equation (GEE) general
linear model (GLM) with outcome as the dependent variable; time in the nursing box, licking / grooming per puppy, vertical nursing per puppy, and ventral nursing per puppy were entered as predictors with breed, maternal parity, sex of puppy, and age at return entered as covariates.
Changes in rates of child diagnoses from baseline to 3 months as a function of mother's remission and subsequently mother's level of response were analyzed using a repeated measures analysis with binary response data, using
generalized estimating equation (GEE) methods.27 A
linear probability
model with an identity link function (rather than a logit - link function) was used to
model interactions on the additive scale28 and to
model a dose - response function using rates (rather than odds) as the outcome measure because we considered risk differences to be a more relevant measure than odds ratios in our study.
Univariate
generalized linear models were used to determine the
estimated marginal means of the PedsQL scales and subscales adjusting for the child's age, sex, maternal education, and disadvantage index as covariates.
We used
generalized linear models with robust variance
estimates to account for within - clinic clustering effects.
Trends in rates of child diagnoses by mother's response level in children with a baseline diagnosis and in rates of incidence or relapse in children without a baseline diagnoses were examined separately using the Cochran - Armitage test for trend.29 Low event rates precluded fitting regression
models adjusting for potential confounders, such as age and sex of child, using
generalized linear models with an identity - link function, to
estimate parameters for adjusted trends.
Generalized regression
models (logistic regression for dichotomous outcomes,
linear regression for continuous outcomes) were used to
estimate the overall adjusted effects of Healthy Steps.26, 27 These
models included site variables to account for the fact that families within sites tend to respond more similarly than those at different sites.
The data was analyzed using
generalized linear models and
generalized estimating equations, which are specifically used to address the multilevel design of data in which schools with participating schoolchildren were randomized (rather than individual participants).