Table 4.
Conditional differences, with their standard errors, and percentage change in estimated difference from unconditional difference in diet quality measures: US adults born between 1951 and 1980 participating in the 2005–2014 National Health and Nutrition Examination Survey
% energy from fat | % energy from saturated fat | Cholesterol (mg) | Fibre (g) | |||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|
β ‡ | se | % Δ§ | β ‡ | se | % Δ§ | β ‡ | se | % Δ§ | β ‡ | se | % Δ§ | |
2007–08 round | −0·16 | 0·47 | −0·18 | 0·18 | 0·86 | 9·96 | −0·41 | 0·72 | ||||
2009–10 round | −1·05* | 0·47 | 11 | −0·66** | 0·18 | 11 | −29·58** | 8·21 | 7 | 1·00 | 0·56 | 24 |
2011–12 round | −1·11* | 0·48 | −14 | −0·79** | 0·19 | 2 | −29·44** | 8·79 | −1 | 2·08** | 0·60 | 6 |
2013–14 round | −0·25 | 0·45 | −0·56** | 0·18 | −4 | −19·97* | 8·20 | 7 | 1·08* | 0·50 | ||
% of energy from FAFH | 0·03** | 0·00 | 0·01** | 0·00 | 0·96** | 0·13 | −0·02** | 0·01 | ||||
% of energy from fast foods | 0·03** | 0·01 | 0·02** | 0·00 | −1·09** | 0·13 | −0·03** | 0·01 | ||||
Constant | 28·71** | 0·87 | 10·61** | 0·43 | 219·01** | 18·20 | 14·05** | 1·17 | ||||
Observations (n) | 12 129 | 12 129 | 12 129 | 12 129 | ||||||||
R 2 | 0·05 | 0·04 | 0·09 | 0·08 |
FAFH, food away from home.
Estimate is statistically significant: *P<0·05, **P<0·01.
Estimated via weighted ordinary least squares with se accounting for the complex sample design. Additional controls include age, household income relative to poverty, household size, and indicators for male, Black, Hispanic, other race/ethnicity (non-Hispanic White is omitted category), married, the dietary intake day was on Friday or on Saturday/Sunday, having completed high school (or General Educational Development) and having at least some college education (no high-school degree is omitted education category).
Calculated only when unconditional difference from 2005–06 is statistically significant. Estimated as [(unconditional – conditional)/unconditional]×100. (Unconditional not shown, but estimable using Table 2.)