mental: Alachua County Study of Mental Impairment

Description

The date set is from a study of mental health for a random sample of 40 adult residents of Alachua County, Florida. Mental impairment is an ordinal response with 4 categories: well, mild symptom formation, moderate symptom formation, and impaired, which are recorded as 1, 2, 3, and 4. Life event index is a composite measure of the number and severity of important life events that occurred with the past three years, e.g., birth of a child, new job, divorce, or death of a family member. It is an integer from 0 to 9. Another covariate is socio-economic status and it is measured as binary: high = 1, low = 0.

Usage

data(mental)

Arguments

Format

mental: Mental impairment. It is an ordinal response with \(4\) categories recorded as \(1\), \(2\), \(3\), and \(4\).
ses: Socio-economic status measured as binary: high = \(1\), low = \(0\).
life: Life event index. It is an integer from \(0\) to \(9\).

References

Agresti, A. (2010) Analysis of Ordinal Categorical Data, 2nd ed. Hoboken, NJ: Wiley.

Examples

Run this code

	# proportional odds model example 
	data(mental)
	
	# model the relationship between the latent variable and life event index as increasing
	# socio-economic status is included as a binary covariate 
	fit.incr <- cgam(mental ~ incr(life) + ses, data = mental, family = Ord)

	# check the estimated probabilities P(mental = k), k = 1, 2, 3, 4
	probs.incr <- fitted(fit.incr)
	head(probs.incr)