How to use ATUS microdata files

Researchers can produce their own time-use estimates using the ATUS microdata files. The ATUS data files include information for nearly 245,000 respondents from 2003 to 2023. Because of the size of these data files, it is easiest to work with them using statistical software such as Stata, SAS, or SPSS.

There are several key pieces of documentation that will aid you when working with the data files. The ATUS User's Guide (PDF) includes information about how to link the data files and use them to produce estimates and standard errors. The Data Dictionaries include variable definitions. The Activity Coding Lexicons provide information about the activity codes. If you would like some hints about useful variables and where to find them, read the ATUS documentation about Frequently Used Variables (PDF).

• About the Basic ATUS Data Files The Basic ATUS data files include data used for generating most time-use estimates.
• About the Additional ATUS Data files The Additional ATUS data files include data about the ATUS survey process, replicate weights for generating standard errors, and other miscellaneous data.
• About the ATUS Eating & Health Module Data files The Eating & Health (EH) Module was fielded from 2006 to 2008, modified slightly, and fielded in 2014 to 2016 and 2022 to 2023. The EH Module data files contain additional information related to eating, meal preparation and health.
• About the ATUS Well-Being Module Data files The Well-Being (WB) Module was fielded for three full years in 2010, 2012, and 2013. It was fielded again from March through December 2021. The WB Module data files contain information related to how people felt during selected activities, as well as general health information.
• About the ATUS Leave Module Data files The Leave Module was fielded from January through December 2011. The 2011 Leave Module data files contain information related to wage and salary workers' access to and use of leave and schedule adjustments. The 2017-18 Leave and Job Flexibilities Module was fielded from January 2017 through December 2018 and included questions about workers’ access to and use of paid and unpaid leave with additional questions about workers' job flexibilities and work schedules.
• Producing Time-Use Estimates
• Linking the ATUS Data files

ATUS data also can be obtained from the American Time Use Survey Data Extract Builder (ATUS-X). (Disclaimer: ATUS-X was not created by BLS and is not maintained by BLS. This link is provided for your convenience and does not constitute an endorsement.)

About the Basic ATUS Data Files:

1. ATUS Respondent File

   This file contains case-specific variables collected in ATUS (that is, variables for which there is one value for each respondent). These include, for example, labor force and earnings information, total time providing secondary childcare, and ATUS statistical weights.

   There is one record for each ATUS respondent.

   Below is a simplified example. The variable TUCASEID identifies each household, and the variable TULINENO identifies each individual within the household. The example contains responses from five individuals; note that the respondent always has TULINENO = 1. In the example, each respondent has a corresponding statistical weight for use in generating estimates representative of the U.S. civilian, noninstitutionalized population (TUFINLWGT), and values for school enrollment (TESCHENR), labor force status (TELFS), and total number of minutes spent alone on the diary day (TRTALONE). The actual ATUS Respondent file contains more variables and records.

   TUCASEID	TULINENO	TUFINLWGT	TESCHENR	TELFS	TRTALONE
   20060101020210	1	22261358.19	1	1	40
   20060101020211	1	5019645.31	1	1	350
   20060101020212	1	2926068.74	1	5	0
   20060101020213	1	25780574.07	2	5	556
   20060101020214	1	3414645.94	1	4	100

    

2. ATUS Roster File

   This file contains information about the age, sex, and relationship to the ATUS respondent of every household member. The same information also is included for the respondent's own non-household children under age 18.

   There is one record for each individual in the respondent's household (including the respondent's own non-household children under age 18).

   A simplified example appears below. The variable TUCASEID identifies each household and the variable TULINENO identifies each individual in the household. In the example below, TUCASEID 20060101020210 has three persons residing in the household, and TUCASEID 20060101020211 and TUCASEID 200601010120212 each have one person. The variable TERRP contains information about the individual's relationship to the ATUS respondent, TESEX shows the person's sex and TEAGE shows the person's age. The actual ATUS Roster file contains more variables and many additional lines.

   TUCASEID	TULINENO	TERRP	TESEX	TEAGE
   20060101020210	1	18	2	42
   20060101020210	2	20	1	45
   20060101020210	3	22	1	11
   20060101020211	1	18	1	65
   20060101020212	1	18	2	21

    

3. ATUS Activity File

   This file includes activity-level information collected in the ATUS, including activity code, location, duration, activity start and stop times, and whether respondents had a child under 13 in their care during the activity.

   There is one record for each activity.

   A simplified example of the ATUS Activity file appears below. This is an illustration of one respondent's day. Because only one person is interviewed per household, each TUCASEID on the Activity file identifies a respondent. Each activity is identified by an activity number (TUACTIVITY_N). The variable TUACTDUR24 shows the duration, in minutes, of each activity during the 24-hour diary day, and the variables TUTIER1CODE, TUTIER2CODE, and TUTIER3CODE correspond to the 1st-, 2nd-, and 3rd-tier of the ATUS activity codes. (The Activity Coding Lexicons provide information about the activity codes.) The ATUS Activity file contains more variables describing each activity as well as many more lines than the example below.

   TUCASEID	TUACTIVITY_N	TUACTDUR24	TUTIER1CODE	TUTIER2CODE	TUTIER3CODE
   20060101020210	1	210	01	01	01
   20060101020210	2	40	13	01	31
   20060101020210	3	40	01	02	01
   20060101020210	4	10	11	01	01
   20060101020210	5	30	18	05	01
   20060101020210	6	180	05	01	01
   20060101020210	7	45	11	01	11
   20060101020210	8	300	05	01	01
   20060101020210	9	30	18	05	01
   20060101020210	10	30	12	01	01
   20060101020210	11	30	11	01	01
   20060101020210	12	180	12	03	07
   20060101020210	13	315	01	01	01

    

4. ATUS Activity Summary File

   The ATUS Activity Summary file contains information about the total number of minutes each respondent spent doing each activity. The file also includes selected variables from the ATUS Respondent, ATUS Roster, and ATUS-CPS files.

   There is one record for each ATUS respondent.

   A simplified example of the ATUS Activity Summary file appears below. The variable TUCASEID is the unique identifier for each respondent and the variable TEAGE, which also appears on the Roster file, shows each respondent's age. The variable t010101 contains the total number of minutes each respondent spent doing activity 010101, "sleeping"; the variable t010102 contains the total number of minutes each respondent spent doing activity 010102, "sleeplessness." The ATUS Activity Summary file contains more variables describing each activity as well as many more lines than the example below.

   TUCASEID	TEAGE	t010101	t010102
   20060101020210	26	480	0
   20060101020211	53	430	30
   20060101020212	76	457	0
   20060101020213	16	600	0

    

5. ATUS Who File

   This file contains codes that indicate who was present during each activity.

   There is one record for each "who" code for each activity. There is one record for activities done alone and multiple records for activities with multiple people present. "Who" codes are not collected for some activities, such as when respondents report sleeping or grooming; in these situations a "-1" denotes that the "who" question was not asked for that activity.

   A simplified example of the ATUS Who file appears below. In the first activity (TUACTIVITY_N = 1), no “who” code information was collected. Only one person was with the respondent during the second activity, so there is one line for TUACTIVITY_N = 2. Three people were with the respondent during the third activity, so there are three lines for TUACTIVITY_N = 3. Two of those present (TUWHO_CODE = 20 and 22) are members of the respondent’s household and can be linked to the Roster file using TULINENO. The third person present (TUWHO_CODE = 51) is not a member of the respondent’s household and thus has a value of "-1" for TULINENO. The actual ATUS Who file contains more variables for each line as well as many additional lines.

   TUCASEID	TUACTIVITY_N	TUWHO_CODE	TULINENO
   20060101020210	1	-1	-1
   20060101020210	2	22	3
   20060101020210	3	20	2
   20060101020210	3	22	3
   20060101020210	3	51	-1

    

6. ATUS-CPS File

   This file contains information collected during the final interview of the Current Population Survey (CPS) for all persons selected to participate in ATUS, whether they did or did not participate, and members of their households. The CPS interviews were conducted two to five months prior to the ATUS interview. For this reason, some of the information on the ATUS-CPS file was out of date at the time of the ATUS interview. When possible, it is best to use variables collected in ATUS rather than their CPS counterparts.

   The file has one record for each household member.

   A simplified example of the ATUS-CPS file appears below. In the example, three households are shown, each represented by a unique value for TUCASEID. Each TULINENO refers to a different household member. TULINENO = 1 refers to the household member who was selected to participate in ATUS. To determine whether this individual participated in ATUS, the variable TRATUSR (starting in 2006), identifies those persons who responded to the ATUS interview. This example shows some of the variables that are not updated in ATUS, such as information about educational attainment (PEEDUCA), race (PTDTRACE), and geographic region (GEREG). The actual ATUS-CPS file contains more variables for each line as well as many additional lines.

   TUCASEID	TULINENO	PEEDUCA	PTDTRACE	GEREG
   20060101020210	1	41	1	4
   20060101020211	1	39	2	3
   20060101020211	2	45	2	3
   20060101020211	3	43	2	3
   20060101020212	1	39	8	1
   20060101020212	2	41	4	1

    

7. ATUS Eldercare Roster File (2011 and later)

   The ATUS Eldercare Roster file contains information about people for whom the respondent provided eldercare. If the respondent indicated that she had provided eldercare more than once during the past 3 to 4 months, additional information about each eldercare recipient is collected. (The time frame varied slightly by respondent because the question asked about care provided between the 1st of a reference month and the interview day.) There is one record for each recipient, up to a maximum of 5 records for each respondent. Information about the relationship of the recipient to the respondent, the age of the recipient, and the duration that care had been provided appear on the file.

   A simplified example of the ATUS Eldercare Roster file appears below. The TUCASEID identifies each respondent providing eldercare, and the TULINENO identifies recipients in the household. A value of -1 for TULINENO indicates that the eldercare recipient does not live in the household. In the example below, TUCASEID 20110101020210 provided care to two persons not living in the household, TUCASEID 20110101020211 provided care to one person, who does live in the household, and TUCASEID 20110101020215 and TUCASEID 20110101020218 each provided care to one person not living in their households. The actual ATUS Eldercare Roster file contains more variables and many additional lines.

   TUCASEID	TULINENO	TEELWHO	TEAGE_EC	TEELDUR
   20110101020210	-1	33	76	4
   20110101020210	-1	34	80	4
   20110101020211	2	20	72	4
   20110101020215	-1	46	88	3
   20110101020218	-1	55	65	2

About the Additional ATUS Data files:

1. ATUS Case History File

   The ATUS Case History file contains case-specific information about the interview process, such as interviewer identifiers, outcome codes, and data quality variables.

   There is one record for each person who was selected to participate in ATUS, whether the selected person did or did not complete an ATUS interview.

   Below is a simplified example. TUCASEID identifies each person selected to participate in ATUS; this example contains information from five individuals who were selected for the ATUS sample. Each value of TUCASEID is associated with variables that list the final outcome code (TRFNLOUT), the number of activities respondents reported doing on their diary day (TUTOTACTNO), the average duration of reported activities (TUAVGDUR), and an interviewer identification number (TUINTID). Note that two of the cases do not show information for TUTOTACTNO and TUAVGDUR; this is because they were noncontacts (that is, they did not participate in ATUS), as is indicated by the value of TRFNLOUT. The actual ATUS Case History file contains additional variables and many more lines.

   TUCASEID	TRFNLOUT	TUTOTACTNO	TUAVGDUR	TUINTID
   20060101020210	001.001	27	42.4	AA01
   20060101020211	001.001	16	90	AA02
   20060101020212	194.001	-1	-1	BB01
   20060101020213	188.001	-1	-1	AA01
   20060101020214	001.001	31	46.5	BB01

    

2. ATUS Call History File (2004 and later)

   The ATUS Call History file contains information about each call attempt that was made to persons selected to participate in ATUS. It includes information such as outcome codes for each call, the date and time of each call attempt, and interviewer identifiers. This file contains information for all persons selected to participate in ATUS, whether they did or did not participate in the survey.

   There is one record for each ATUS call attempt.

   Below is a simplified example. TUCASEID identifies each ATUS case and TUATTMPTNO identifies each call attempt. The example shows information from two cases that were selected for the ATUS sample. The first case (TUCASEID = 20060101020210) received two calls from an ATUS interviewer, with the second call resulting in a complete interview; the second case (TUCASEID = 20060101020211) received three calls before resulting in a complete interview. The variable TRFNLCLL shows the outcome of each call. An anonymous interviewer identification number (TUHINTID) is associated with each call attempt.

   TUCASEID	TUATTMPTNO	TUATTMDATE	TRFNLCLL	TUHINTID
   20060101020210	1	20060102	188.001	AA01
   20060101020210	2	20060102	001.001	BB01
   20060101020211	1	20060103	188.001	BB01
   20060101020211	2	20060103	188.001	AA01
   20060101020211	3	20060110	001.001	BB01

    

3. ATUS Trips File (years 2005-10 only)

   The Trips file contains information about the number of times each respondent was away from home for 2 nights or more in a specific reference month. This file does not include weights.

   For more information about the Trips file read the Notice about Trips file.

    

4. ATUS Replicate Weights File

   The ATUS Replicate Weights file contains weights necessary for generating standard errors for ATUS estimates. The file contains one record for each individual selected to participate in ATUS, whether the person did or did not participate in the survey. Technical information about the Replicate weights file can be found in the text document that is enclosed in the Replicate weights zip file. The formula for calculating standard errors with the replicate weights can be found in Chapter 7 of the ATUS User's Guide (PDF).

    

5. ATUS Pandemic Replicate Weights File (years 2019-2020 only)

   The ATUS Pandemic Replicate weights file for 2019-20 contains 160 replicate final weights for each ATUS final weight created using the 2020 weighting method. See Chapter 7 of the ATUS User's Guide (PDF) for more information about the 2020 weighting method..

About the ATUS Eating & Health Module Data files:

1. EH Respondent File

2. The EH Respondent file contains case-specific variables collected in the EH Module (that is, variables for which there is one value for each respondent). These include, for example, general health information, body mass index values, and EH Module statistical weights.

   There is one record for each EH Module respondent.

   Below is a simplified example. The TUCASEID identifies each household, and TULINENO identifies each individual within the household. The example contains responses from five individuals; note that the respondent always has TULINENO=1. In the example, each respondent has a corresponding statistical weight (EUFINLWGT) for use in generating estimates representative for the U.S. civilian, noninstitutionalized population age 15 and over. Estimates using EH Module data should be generated using EUFINLWGT instead of the ATUS statistical weight (TUFINLWGT) because there are fewer EH Module respondents than ATUS respondents due to minimal nonresponse by ATUS respondents to the EH Module. This example also demonstrates that each respondent has corresponding values denoting general health status (EUGENHTH) and total time spent in primary eating and drinking (ERTPREAT). The actual EH Respondent file contains many more variables as well as many more lines.

   TUCASEID	TULINENO	EUFINLWGT	EUGENHTH	ERTPREAT
   20060101020210	1	23264358.19	1	40
   20060101020211	1	7013642.31	5	350
   20060101020212	1	5780754.07	2	556
   20060101020213	1	18426068.74	3	0
   20060101020214	1	2916445.94	1	100

    

3. EH Child File (2006-2008 only)

   The EH CHild file contains information about which household children under age 19 ate a breakfast or lunch prepared at a school, day care, Head Start center, or summer day program.

   If at least one child in the respondent’s household ate such a breakfast or lunch, then there is one record for each child under age 19 in the respondent’s household, regardless of whether or not each child ate such a breakfast or lunch. For example, a household with three children under 19, only one of whom ate a school meal, will have a record for all three children on EH Child. Information about children under age 19 in households where no one ate a prepared meal will not be on this file.

   A simplified example appears below. TUCASEID identifies each household, and TULINENO identifies each household child under the age of 19. In the example below, TUCASEID 20060101020210 has two children under age 19. The first (TULINENO=2) ate a prepared breakfast (EEBRK=1) and did not eat a prepared lunch (EELCH=2). The second child (TULINENO=3) did not eat a prepared breakfast or a prepared lunch. The actual EH Child file contains many more variables and many additional lines.

   TUCASEID	TUACTIVITY_N	EEBRK	EELCH
   20060101020210	2	1	2
   20060101020210	3	2	2
   20060101020211	3	1	1
   20060101020211	4	1	1
   20060101020212	2	2	1

    

4. EH Activity File

   The EH Activity File includes activity-level information collected in the EH Module, including activity number, whether secondary eating occurred during the activity, and duration of secondary eating.

   There is one record for each activity.

   A simplified example of the EH Activity file appears below. This is an illustration of one respondent’s day. Because only one person is interviewed per household, each TUCASEID on the EH Activity file identifies a respondent. Each activity is identified by an activity number (TUACTIVITY_N). In the example, each activity has corresponding values denoting whether secondary eating occurred during the activity (EUEATSUM) and how much time was spent in secondary eating (EUEDUR24). The actual EH Activity file contains more variables describing each activity as well as many more lines than does the example below.

   TUCASEID	TUACTIVITY_N	EUEATSUM	EUEDUR24
   20060101020210	1	-1	-1
   20060101020210	2	1	20
   20060101020210	3	1	10
   20060101020210	4	-1	-1
   20060101020210	5	-1	-1

    

5. EH Replicate Weights File

   The EH Replicate Weights file contains weights necessary for generating standard errors for EH Module estimates. The file contains one record for each individual who responded to the EH Module. Technical information about the EH Replicate weights file can be found in the text document that is enclosed in the EH Module Replicate Weights zip file. The formula for calculating standard errors with the replicate weights can be found in Chapter 7 of the ATUS User's Guide (PDF).

About the ATUS Well-Being Module Data files:

1. WB Respondent File

2. The WB Respondent file contains case-specific variables collected in the WB Module (that is, variables for which there is one value for each respondent). These include, for example, general health information, how well-rested the respondent felt on his diary day (the day about which he was interviewed), and WB Module respondent-level statistical weights.

   There is one record for each WB Module respondent.

   Below is a simplified example. The TUCASEID identifies each household, and TULINENO identifies each individual within the household. The example contains responses from 5 individuals; note that the respondent always has TULINENO=1. (All records on the WB Respondent file have TULINENO=1 because only one person in the household responded to the ATUS.) In the example, each respondent has a corresponding statistical weight (WUFINLWGT) for use in generating estimates. The respondent-level weight WUFINLWGT is similar to the ATUS statistical weight TUFINLWGT except that it is adjusted to account for minimal nonresponse by ATUS respondents to the WB Module. WUFINLWGT should be used instead of TUFINLWGT for many estimates derived using the Well-being Module data; an exception is when the calculation uses module data about the randomly-selected activities appearing on the WB Activity file. This example also demonstrates that each respondent has corresponding values denoting general health status (WEGENHTH) and how well-rested the respondent felt (WEREST). The actual WB Respondent file contains additional variables as well as many more lines.

   TUCASEID	TULINENO	WUFINLWGT	WEGENHTH	WEREST
   20100101100605	1	4040229.08	5	2
   20100604101229	1	1438440.99	4	1
   20100807100994	1	3151945.82	2	4
   20060101020213	1	18426068.74	3	0
   20100908101868	1	1621125.07	1	3
   20101211101129	1	14724370.75	3	1

    

3. WB Activity File

4. In the WB Module, the survey randomly selected 3 activities reported by each respondent. For each selected activity, the respondents were asked 7 questions related to quality of life (5 affect questions, 1 question about how meaningful the activity was, and 1 question about whether the respondent was interacting with anyone during the activity). The affect questions asked how happy, tired, stressed, sad, and in pain the respondent felt during the activity. The order of the 5 affect questions was randomly determined for each respondent. This file includes the activity-level information collected in the WB Module, including activity number, and information about how the respondent felt during the selected activities.

   There is one record for each activity selected for the module. The activity must have been at least 5 minutes in duration to have been selected for the module. The following activities and responses were not eligible for selection:

   • Sleeping (0101xx)
   • Grooming (0102xx)
   • Personal Activities (0104xx)
   • Don’t know/Can’t remember (500106)
   • Refusal/None of your business (500105)
   
   
   

   A simplified example of the WB Activity file appears below. This is an illustration of one respondent’s day. Because only one person was interviewed per household, each TUCASEID on the WB Activity file identifies a respondent. Each activity is identified by an activity number (TUACTIVITY_N). In the example, each activity has corresponding values denoting how happy the respondent felt during the activity (WUHAPPY), whether the respondent was interacting with anyone during the activity (WUINTERACT), and the activity-level statistical weight (WUFNACTWTC). The actual WB Activity file contains more variables describing the feelings associated with each selected activity as well as many more lines than does the example below.

   TUCASEID	TUACTIVITY_N	WUHAPPY	WUINTERACT	WUFNACTWTC
   20100101102152	2	3	2	12059879.6
   20100101102152	7	1	2	2009974.4
   20100101102152	13	6	1	4019965.4

   WB Activity Weights

   The WB Module activity weight variables, WUFNACTWTC on the 2010, 2012, and 2013 single-year data files, WUFNACTWT on the 2021 data file, and WUFNACTWTP on the multi-year data file, are designed for use with data collected about the randomly-selected activities in the module. WUFNACTWT (2021), WUFNACTWTC (2010, 2012, and 2013), or WUFNACTWTP can be used to estimate average levels of affect for the population during activities eligible for the module. The weights account for differences between activities in the fraction of time spent in eligible activities. They also account for differences between persons in the probability of having a specific eligible activity selected due to variation in the number of eligible activities. For more information about the WB Module activity weight variables, see the ATUS WB Module Data Dictionary.

   NOTE: There was an error in the activity selection process for the 2010, 2012, and part of the 2013 Well-Being Module to the ATUS. The adjusted activity weights (WUFNACTWTC on the single-year data files and WUFNACTWTP on the multi-year data file) help mitigate this error. Read More

    

5. WB Replicate Weights Files

6. The WB Replicate Weights files contain weights necessary for generating standard errors for WB Module estimates. The WB Respondent Replicate weights file contains one record for each individual who responded to the WB Module. The WB Activity Replicate weights file contains one record for each activity selected for the WB Module. Technical information about the WB Replicate weights files can be found in the text document that is enclosed in the WB Module Replicate Weights zip files. The formula for calculating standard errors with the replicate weights can be found in Chapter 7 of the ATUS User's Guide (PDF).

   NOTE: There was an error in the activity selection process for the 2010, 2012, and part of the 2013 Well-Being Module to the ATUS. The adjusted activity replicate weights (WUFNACTWTC on the single-year data files and WUFNACTWTP on the multi-year data file) help mitigate this error. Read More

About the ATUS Leave Module Data files:

1. LV Respondent File

2. This file contains case-specific variables collected in the Leave Module (that is, variables for which there is one value for each respondent). These include, for example, information about access to paid leave, whether the respondent took paid leave during the past 7 days, and Leave Module statistical weights.

   There is one record for each Leave Module respondent.

   Below is a simplified example. The TUCASEID identifies each household, and TULINENO identifies each individual within the household. The example contains responses from 6 individuals; note that the respondent always has TULINENO=1. (All records on the Leave Respondent file have TULINENO=1 because only one person in the household responded to the ATUS.) In the example, each respondent has a corresponding statistical weight (LUFINLWGT) for use in generating estimates representative of wage and salary workers in the U.S. civilian noninstitutional population age 15 and over. The statistical weight LUFINLWGT should be used instead of the ATUS statistical weight TUFINLWGT when using Leave Module data. There are fewer Leave Module respondents than ATUS respondents because the module was asked of wage and salary workers only, and some eligible ATUS respondents did not complete the module; LUFINLWGT accounts for the difference in the ATUS and Leave Module populations and for minimal nonresponse.

   This example also demonstrates that each respondent has corresponding values denoting access to paid leave (LUPAID) and whether the respondent took leave during the past 7 days (LULEAVE). The actual LV Respondent file contains additional variables as well as many more lines.

   TUCASEID	TULINENO	LUFINLWGT	LUPAID	LULEAVE
   20110101100702	1	5620229.08	1	2
   20110604101236	1	1438663.25	2	1
   20110807100989	1	3151789.63	2	2
   20110101020213	1	56896068.74	1	2
   20110908142868	1	1628924.07	1	1
   20111215623129	1	14895670.75	2	1

    

3. LV Replicate Weights File

   The LV Replicate Weights file contains statistical weights necessary for generating standard errors for Leave Module estimates. The LV Replicate Weights file contains one record for each individual who responded to the Leave Module. Technical information about the LV Replicate Weights file can be found in the text document that is enclosed in the LV Replicate Weights zip file. The formula for calculating standard errors with the replicate weights can be found in Chapter 7 of the ATUS User's Guide (PDF).

    

Producing Time-Use Estimates:

Researchers often are interested in estimates of the amount of time Americans spend doing a specific activity, such as the average time Americans watched television in 2003. This type of estimate can be generated by using the Basic ATUS data files. The simplest way to generate an estimate about time use on an average day involves using the ATUS Activity Summary file. Researchers who wish to investigate more complicated questions, such as the amount of time Americans spent watching TV with a family member, will need to link multiple files. To answer this particular question, researchers would need to link the ATUS Respondent file (to obtain the statistical weight), the ATUS Activity file (to obtain activity information and duration), and the ATUS Who file (to obtain information about who was with the respondent during each activity). Those generating estimates using the module files in addition to the basic or additional ATUS files should use the module weights. For more information about how to calculate estimates, see Chapter 7 and Appendix J of the ATUS User's Guide (PDF).

Linking the ATUS Data files:

Each of the ATUS microdata files contains useful information, but to produce most estimates, the files must be linked. All of the microdata files contain the variable TUCASEID, which is the ATUS identification number. TUACTIVITY_N (the activity line number) and TULINENO (the person line number) are two additional linking variables that can be used in conjunction with TUCASEID.

1. BASIC ATUS DATA FILES

2. File	Linking Variables
   Respondent file	TUCASEID TULINENO (always equal to 1 on the Respondent file)
   Roster file	TUCASEID TULINENO
   Activity file	TUCASEID TUACTIVITY_N
   Who file	TUCASEID TUACTIVITY_N TULINENO
   ATUS-CPS file	TUCASEID TULINENO
   Activity summary file	TUCASEID
   Eldercare roster file (2011 and later)	TUCASEID TULINENO

    

3. ADDITIONAL ATUS DATA FILES

   File	Linking Variables
   Case history file	TUCASEID
   Call history file (2004 and later)	TUCASEID
   Trips file (2005 to 2011)	TUCASEID
   Replicate weights file	TUCASEID
   Pandemic Replicate weights file (2019-2020)	TUCASEID

    

4. ATUS EH MODULE FILES

   File	Linking Variables
   EH Respondent file	TUCASEID TULINENO
   EH Child file (2006-2008)	TUCASEID TULINENO
   EH Activity file	TUCASEID TUACTIVITY_N
   EH Replicate weights file	TUCASEID

    

5. ATUS WB MODULE FILES

   File	Linking Variables
   WB Respondent file	TUCASEID TULINENO
   WB Activity file	TUCASEID TUACTIVITY_N
   WB Respondent Replicate weights file	TUCASEID
   WB Activity Replicate weights file	TUCASEID TUACTIVITY_N

    

6. ATUS LEAVE MODULE FILES

   File	Linking Variables
   LV Respondent file	TUCASEID TULINENO
   LV Replicate weights file	TUCASEID

    

The ATUS files can also be linked to CPS files. More information is available in Appendix K of the ATUS User’s Guide.

Last Modified Date: June 27, 2024