Data Processing & Data Analysis

Data Processing Requirements

Data Analysis Considerations

For greater detail on the following issues when considering whether to use a 24HR to answer a particular research question, see Choosing an Approach for Dietary Assessment.

General Considerations

Guidance for Specific Research Objectives

  • If your objective is to estimate solely the mean intakes of a group, defining the mean for a proportion or [glossary term:] ratio appropriate to the particular objective requires further consideration. For example, data from multiple repeated administrations of 24HRs can be used to estimate average per person or population ratios and proportions (Learn More about Ratios and Proportions). In addition, for studies that have collected multiple 24HRs, it is unnecessary to adjust for within-person random error.
  • If your objective is to estimate [glossary term:] usual dietary intake distributions for a group (e.g., to examine percentiles or to estimate the proportion above or below some threshold), clarification of whether the focus is on habitual intake over the long run or intake on a given day (i.e., acute intake) is required (Learn More about Usual Dietary Intake).
  • If your research objective is to analyze the [glossary term:] association between diet as an [glossary term:] independent variable and another variable (e.g., between diet at baseline and onset of cancer), and the 24HR is the main instrument, statistical modeling of data from multiple administrations of the 24HRs will account for within-person random error, allowing greater [glossary term:] precision in the estimates of associations, and thus increasing statistical [glossary term:] power.
  • If your research objective is to analyze the association of an independent variable (e.g., socioeconomic status) and diet as the dependent variable, statistical modeling to remove within-person random erro is not necessary. However, averaging 24HRs across multiple days may increase the precision of the diet estimate and thus the statistical power to detect associations. In addition, variables known to affect quality of report (e.g., body mass index) should be included as [glossary term:] covariates in analyses.
  • If your research objective is to analyze a change in diet as a result of an intervention, the potential for [glossary term:] differential response bias must be considered. To avoid the effects of this potential bias, an objective measure, such as measurement of serum carotenoids as a marker for fruit and vegetable intake, could be considered. In such a case, a 24HR may be used as a secondary source of information.
    • In intervention studies in which an objective instrument, such as a [glossary term:] biomarker, has been used as the main instrument and 24HRs have been used as a secondary instrument, the agreement between the objective instrument and the 24HRs should be examined. If there is substantial agreement, the 24HRs may be useful for analyses of additional dietary factors beyond those measured by the objective instrument.
    • In intervention studies in which 24HRs are used as the main instrument, variables known to be related to reporting [glossary term:] accuracy (e.g., body mass index) and to differential response bias (e.g., social desirability score) should be included as covariates.
    • Respondent burden may cause [glossary term:] attrition. The extent and nature of this attrition must be considered in the analyses, for example, by comparing the objective measures and the self-reported diets at baseline of those who completed the study to those who dropped out.