Analysis of paired Likert data: how to evaluate change and preference questions.

BACKGROUND AND OBJECTIVES: Likert scale data present unique analysis concerns that are often not recognized by nonstatistical researchers. Mean responses to a single question are frequently compared among two or more groups of moderate sample size using normal theory methods (t test, ANOVA). These tests are designed for continuous normally distributed data, but Likert responses are categorical, ordinal, and not normally distributed. The problems are further complicated when the paired nature of pre/post- or preference questioning is ignored. METHODS: Simulated data for a single five-point Likert scale question was used to illustrate the differing conclusions that may arise from single-question Likert scale data depending on whether pairing is modeled appropriately and which statistical procedure was applied (two sample: t test or Wilcoxon rank sum; paired samples; t test, Wilcoxon signed rank, or sign test). RESULTS: Cases are shown for which the paired responses are significantly different (indicated by the sign test), although the group mean responses to the two questions are identical (leading to nonsignificance with both paired t and signed rank tests). The sign test is shown to be appropriate and simple to apply in typical pre/post- or preference situations. CONCLUSIONS: Correctly identifying the question of interest is critical to selecting the appropriate statistical methodology.

Home environment and adaptive social behavior among premature, low birth weight children: alternative models of environmental action.

Used data from 465 premature, low birth weight children representing three major sociocultural groups (Caucasian, African American, Hispanic) to examine the relation between children's home environments and their adaptive social behavior. Results showed low to moderate associations between scores on the HOME Inventory at 1 and 3 years and scores on two measures of adaptive social behavior at 30 to 36 months, the Adaptive Social Behavior Inventory, and observations of mother-child interaction in a structured laboratory situation. Results indicated that responsive, nurturant care at both 1 and 3 years are related to child adaptive social behavior, as are cognitively stimulating experiences and materials. However, canonical correlational analysis indicated that only Acceptance and Variety of Experience, measured at age 3, and Variety of Experience measured at age 1 accounted for independent amounts of variance in adaptive social behavior as perceived by mothers. Also, only sociocultural group status and Learning Materials at 36 months contributed to the prediction of persistence and enthusiasm as observed in the laboratory setting.

A reexamination of the association between HOME scores and income.

This study, which represents another look at the relationship between the HOME Inventory and income, uses data from the Infant Health and Development Program (IHDP), a multisite, longitudinal study of low-birth-weight preterm infants. Two versions of the HOME Inventory were used: The Infant/Toddler (IT-HOME), at 12 months of age, and the Early Childhood (EC-HOME), at 36 months of age. Predictor variables were income, ethnicity, maternal education, parity, gestational age, marital status, maternal age, and site. HOME scores were positively correlated with income. However, after controlling for the other variables in the models, the portion of the variance in HOME scores uniquely explained by income was quite low (IT-HOME, 5.1%; EC-HOME, 4.2%). Finally, the relationship between HOME scores and four child characteristics (cognitive development, growth, maladaptive behavior, and social competence) measured when the child was 36 months old were investigated using correlation. The results indicated that the quality of the home environment, as measured by the HOME Inventory, is related to children's development.

A factor analytic study of the infant-toddler and early childhood versions of the HOME Inventory administered to white, black, and Hispanic american parents of children born preterm.

Factor analyses were performed on the Infant-Toddler and the Early Childhood versions of the HOME Inventory for 3 groups (blacks, whites, and Hispanics) of premature, low-birthweight children. Participants lived in 8 different U.S. cities. On the IT-HOME, 5 factors were originally retained for each group using the principal factors method of extraction. Factor structures for blacks and whites were similar, accounting for 85% and 75% of the variance, respectively. The structure for Hispanics was somewhat different; a 7-factor solution accounting for 65% of the variance was most interpretable. For the EC-HOME, 6 factors were retained for both black and white groups, accounting for 80% and 73% of the variance, respectively. For Hispanics, an 8-factor solution accounted for 59% of the variance. The factor structures for blacks and whites were largely in agreement with the current organization of items into subscales. The fit for Hispanics was not quite as good, but most factors also corresponded to current subscales.

Early indications of resilience and their relation to experiences in the home environments of low birthweight, premature children living in poverty.

The caregiving environment experienced by 243 premature, low birthweight (LBW) children living in poverty was examined to determine whether the quality of care such children receive affords them some measure of protection from the generally deleterious consequences of poverty and prematurity. Only 26 children were identified as functioning in the normal range for cognitive, social/adaptive, health, and growth parameters at age 3. These children, who showed early signs of resiliency, differed from nonresilient children in that they were receiving more responsive, accepting, stimulating, and organized care. They were also living in safer, less crowded homes. 6 "protective" aspects of caregiving were identified and used as part of a cumulative protection index. Children with less than 3 protective aspects of caregiving present at age 1 had only a 2% probability of being resilient, and only a 6% probability if fewer than 3 were present at age 3. Overall, premature LBW children born into conditions of poverty have a very poor prognosis of functioning within normal ranges across all the dimensions of health and development assessed. However, those raised in a setting with 3 or more protective factors were more likely to show early signs of resiliency.