SED_EAT is a 4-digit numeric variable.
9996 = Refused
9997 = Don't know
9999 = NIU (Not in Universe)
SED_EAT is the sum of all time during the diary day that the respondent spent in secondary eating. Secondary eating is eating that occurred during a primary activity.
Secondary eating cannot occur during the following primary activities: 010101 (sleeping), 050202 (eating and drinking as part of job), 110101 (eating and drinking), 110199 (eating and drinking, n.e.c.), and 119999 (eating and drinking, n.e.c.).
Researchers may also want to use SED_DRINK, which reports the total amount of time in secondary drinking. However, users should keep in mind that SED_EAT and SED_DRINK should not be added to create a total of secondary time in eating and drinking as there may be some overlap in specific episodes of secondary eating and drinking (see additional documentation). Researchers may also want to use the secondary eating and drinking filter in the extract builder, or use the total amount of time in primary eating and drinking (PED).
Total time in secondary eating was calculated based on information collected during the Eating and Health Module of the ATUS interview. Analyses that include SED_EAT should use EHWT as the weight.
This variable is only available as part of the Eating and Health Module.
In October 2006, an "all day" option was added for secondary eating. For respondents who reported "all day," the secondary eating activity duration is equivalent to the duration of the primary activity. Secondary eating is not allowed during a small number of primary activities, including sleeping (010101), primary eating and drinking (110101) eating and drinking, n.e.c. (110199, 119999), and eating and drinking as part of job (050202).
Respondents were able to report "all day" prior to October 2006, and the mean duration of secondary eating increased. The percentage of respondents reporting any secondary eating remained fairly level over 2007. No internal consistency issues exist for 2006 and 2007. Pooling 2006 and 2007 data should take the skewness of the data into account.
- Eating and Health module respondents.
- 2006-2008, 2014-2016