Time use: documentation of statistics

The Time Use Survey is a sample survey, the participants of which keep a detailed record of their time use for two days. In addition to the time spent on various activities, the survey investigates the daily and weekly rhythms of time use and time spent together with others. The survey looks at working hours, the time spent on domestic work, sleeping and meals, various leisure pursuits and the time spent together with others and the location of activities. The rhythms of time use are also examined according to a 24-hour period, the day of the week and the season. The activities written about by respondents in their own words were coded according to a 161-category activities classification from 2020 to 2021.

The Time Use Survey uses a household sample, in which the survey units are households and individuals, who are aged 10 and over at the time of the survey. The sampling design is a one-stage cluster sampling. The sample of 2020 to 2021 was drawn from Statistics Finland’s statistical register pertaining to the population which had been adapted from the Central Population Centre.
The register does not contain data on households, due to which household-dwelling units, instead of households, are used in the sampling. A household-dwelling unit was formed of individuals with an identical address code in the register. In most cases, one household-dwelling unit is equal to a single household, due to which the household-dwelling unit sample was used to target the data collection at households. The survey’s statistical population is the household population, meaning that individuals living in institutions and conscripts were removed from the sample as falling under the scope of the register’s over-coverage.

The Time Use Survey’s statistical population is the household population aged 10 and over. Household members aged 10 and over are interviewed and provided with a time-use diary to fill in.  The respondents keep a record of their time use over two 24-hour periods, one of which is a weekday and the other a weekend day. In addition, those employed keep a record of their working hours for seven consecutive days.

The diaries provide data describing the duration, timing and sequence of the activities. The durations of activities are usually indicated as averages (hours and minutes per a 24-hour period) and timing as percentages pertaining to different times of the day or as graphic tempograms. The durations of activities can be calculated as averages pertaining to an entire population group, in which case the calculations include all respondents, regardless of whether they participate in the activities in question. The share of individuals within a population group who engaged in a particular activity during the survey day (‘participation percentage’) can also be calculated. An average can also be calculated for those participating in an activity (‘time among participants’).

September and November 1979, April 1987 to March 1988, March 1999 to February 2000, April 2009 to May 2010, September 2020 to September 2021.

The data covers the entire country. The area-level classifications are: statistical municipality grouping, major region and region.

The Time Use Survey’s statistical population is the household population aged 10 and over. The data of the survey cover time use over an entire year. The data collection lasts for 365 consecutive days. The survey year is not necessarily a calendar year. The time use data are collected to the nearest ten minutes over an entire 24-hour period.

Statistics Finland has conducted countrywide Time Use Surveys five times, at roughly 10-year intervals. The latest survey was conducted from 2020 to 2021. The previous surveys were conducted in 1979, from 1987 to 1988, from 1999 to 2000 and from 2009 to 2010. The data of the first survey were collected in September and November, and concerned the population aged between 10 and 64. The subsequent studies were conducted over a 12-month period, and concerned the population aged 10 and over.

New data are released approximately once a decade. The material concerning a single year is published in several releases covering different areas of content.

The Time Use Survey contains four different types of non-response: 1) household non-response, 2) individual non-response, 3) diary non-response, in which either both or one of the diaries is missing, and 4) item non-response, in which the response to some questions is missing or a diary is only partially completed.  The volume of elementary unit non-response is measured by indicating the share of elementary units remaining non-responsive as percentages. If all members of a household participate in activities, and time is spent together, there is intraclass correlation in the research variables in relation to the household. Individuals may have habits which are identical from one day to the next.

The design coefficients calculated for the variables allow for estimating the clustering of the variables in the household sample.

The publication of the results of the Time Use Survey conducted from 2020 to 2021 began in June 2022.

New data have been released on schedule, in accordance with the release plan.

Revisions – i.e. improvements in the accuracy of statistical data already published – are a normal feature of statistical production and result in improved quality of statistics. The principle is that statistical data are based on the best available data and information concerning the statistical phenomenon. On the other hand, the revisions are communicated as transparently as possible in advance. Advance communication ensures that the users can prepare for the data revisions. 

The reason why data in statistical releases become revised is often caused by the data becoming supplemented. Then the new, revised statistical figure is based on a wider information basis and describes the phenomenon more accurately than before. 

Revisions of statistical data may also be caused by the calculation method used, such as annual benchmarking or updating of weight structures. Changes of base years and used classifications may also cause revisions to data. 

Content-wise, the data concerning Finland in the statistics are geographically comparable. Comparisons can be done within the framework set by the scope of the data. As was the case with the surveys conducted from 1999 to 2000 and from 2009 to 2010, the Time Use Survey conducted from 2020 to 2021 was EU-harmonised. Some 20 European countries conducted a harmonised Time Use Survey for the first time at the turn of the century. The same number of countries conducted the survey between 2008 and 2015. Between 2019 and 2025 over 20 European countries will conduct the survey. Eurostat's guidelines allow the use of nationally compiled interview questions and the addition of national time use categories to the basic classification frame. However, the format of the diary used is fully harmonised.

At a rough level, the time use classification can be restored to 26 temporally comparable categories back to 1979, to 82 categories back to the years 1987 and 1988, to 132 categories back to years 1999 and 2000, and to 158 categories back to years 2009 and 2010. In addition, individual activities can be compared to each other. The 1979 data differ from the data of other years in the sense that, rather than covering an entire year, they cover only the autumn. In addition, social conditions during the time of data collection may have an impact on the results pertaining to time use, for example.

In terms of gainful employment, time use can be compared to the quarterly and annual results of Statistics Finland’s Labour Force Survey. The times are not precisely overlapping, however.

Both the Time Use Survey and the Leisure Survey produce data on participation in cultural events. Comparable data are available as of 1981. The differences in data collection must be accounted for when comparing the data of the Time Use Survey to those of the Leisure Survey. The questions pertaining to participation in cultural events in the Time Use Survey were phrased in a manner consistent with the questions of the Leisure Survey. The results of the Time Use Survey and the Leisure Survey are fairly coherent in general. Given that the Time Use and Leisure Surveys are conducted at alternating time intervals, comparable time series can be formed.

The questions pertaining to trust in the Time Use Survey have been phrased in a manner consistent with the questions of the Leisure Survey.

The time use diaries are well suited for surveying daily activities. These include sleeping, meals, washing and many other activities related to personal care, as well as many domestic chores, such as cooking and childcare. Interviews are a suitable method for complementing data on more infrequent activities. Different methods yield different figures on participation in organisations, for example. The definitions and reference periods on which the results are based usually differ from each other.

The Time Use Survey uses a household sample, in which the survey units are households and individuals, who are aged 10 and over at the time of the survey. The sampling design is a one-stage cluster sampling. The sample was drawn
from Statistics Finland’s statistical register pertaining to the population which had been adapted from the Central Population Centre. The register does not contain data on households, due to which household-dwelling units, instead of households, are used in the sampling. A household-dwelling unit was formed of individuals with an identical address code in the register. In most cases, one household-dwelling unit is equal to a single household, due to which the household-dwelling unit sample was used to target the data collection at households.

The households drawn for the sample of the Time Use Survey conducted from 2020 to 2021 numbered 8,840. These included 21,278 individuals, of whom 19,268 was aged 10 and over. The household sample was drawn as an individual sample in such a way that individuals aged 15 or over, for whom individuals living in the same dwelling unit were simultaneously included, were drawn from the register by systematic sampling. The likelihood of a household being drawn is dependent on the number of the household’s individuals falling under the scope of the sampling frame, i.e. in this case, the number of individuals aged 15 or over. The survey’s statistical population is the household population, meaning that individuals living in institutions and conscripts were removed from the sample as falling under the scope of the register’s over-coverage. The households belonging to over-coverage numbered 185 (2.1 per cent). Following their removal, the sample included 8,655 households. There were 4,100 households which responded; this represents 47.4 per cent of the sample.

The data of the Time Use Survey conducted from 2020 to 2021 were collected with computer-assisted telephone interviews (CATI) and with web and paper diaries. Household members aged 10 and over kept a record of their time use over one weekday and one weekend day. Those who work also kept a weekly diary on their gainful employment for a period of seven days. Data on employment, studying, voluntary work and free time hobbies were inquired about in the survey’s interview part.

In the previous Time Use Surveys, in 1979, from 1987 to 1988 and from 1999 to 2000, the data were collected during face-to-face interviews, but in the survey from 2009 to 2010, some of the interviews were carried out as telephone interviews in the interest of cutting costs. In the survey from 2020 to 2021 all of the interviews were carried out as telephone interviews.

The data collection takes place roughly once in every 10 years. The data collection lasts a year.

In some diaries, an activity at some point during a day had not been entered. If the diary had been kept for most of a day, the diary was accepted in the data, and the missing data were given the code 999 as a value for the activity category. For some
of the 999 codes, the time spent on sleeping could be predicted with the help of a model.  If the keeping of the diary had been started in the morning with morning routines, the time from 4.00 until the morning routines was marked as sleeping. Sleeping that took place in the evening was modelled with a regression model in which the dependent variable was the duration of the evening sleeping and the independent variables were sleeping in the morning, sex, five-year age groups, the day of the week, working day and day off. The model predicted the duration of the evening sleeping, which was rounded to the nearest 10 minutes. If the final episode was coded as missing, it was imputed with the evening sleeping calculated with the model. If the final missing episode was shorter than the sleeping predicted by the model, the entire missing amount of time was imputed to sleeping. If, on the other hand, the duration of the final episode was greater than the sleeping predicted by the model, the duration calculated with the model was imputed as the duration. The difference between the durations of the model and the missing data remained as a missing activity. The time at which the sleeping started was calculated retroactively as of 4.00.
Weights are calculated at several stages which account for the survey and sampling design and the correction of the non-response impact. The weights are also standardised with a calibration technique to correspond to data obtained from population statistics and registers. The weighting gives days of the week and months the same ‘representativeness’ in the sense that the sums of the weights calculated from a unit level are the same.
A household weight was calculated for a household and all members of a household were given the same weight, in accordance with the sampling design. This weight was used when calculating the shares of respondents and the distributions of non-response.
The calibration vector was formed according to age and gender groups, the area, municipality type, education and taxable income. The standardisation of a month and the day of the week, as well as inclusion in the register of unemployed job seekers during the survey month, were linked to same calibration. 
Not all members of a household kept a diary or kept it only for one day, due to which the weight based on the inclusion probability is expanded to apply to the survey’s statistical population, i.e. household population aged 10 and over. The basic weight of the diaries was 5/7 for weekdays and 2/7 for days of the weekend. The numbers of diaries received varied depending on the survey week and day of the week. If an individual was not reached before a day drawn by lot, the interviewer had the possibility to postpone the keeping of the diary for no further than the next three weeks and the same days of the week.  Due to such postponements, not all members of a household have the same diary weights. If all members of a household kept their diaries on the same days, the members of the household have the same diary weights. Day weights were calibrated in a manner equivalent to individual weights.

The data are tested as soon as parts of it are completed. The results are compared to those of the earlier surveys.
In the interview data, the Blaise program accepts only valid codes. The interview data are checked by carrying out logicality checks and by comparing the distributions to earlier Time Use Surveys and other sources, such as the Labour Force Survey, the Leisure Survey and the survey on the use of information and communications technology by individuals.
The application in which the diary data are stored accepts only valid codes. That the sum of the continuous variables is 1,440 minutes on every survey day is verified. It is also checked whether the keeping of a diary has been interrupted, in which case a missing evening napping is potentially imputed. The averages of continuous variables are compared with the averages of the previous survey.

Statistics Finland

Social Statistics

The compilation of statistics is guided by the Statistics Act. The Statistics Act contains provisions on collection of data, processing of data and the obligation to provide data. Besides the Statistics Act, the Data Protection Act and the Act on the Openness of Government Activities are applied to processing of data when producing statistics. 

Statistics Finland compiles statistics in line with the EU’s regulations applicable to statistics, which steer the statistical agencies of all EU Member States.  

Further information: Statistical legislation 

The data protection of data collected for statistical purposes is guaranteed in accordance with the requirements of the Statistics Act (280/2004), the Act on the Openness of Government Activities (621/1999), the EU's General Data Protection Regulation (EU) 2016/679 and the Data Protection Act (1050/2018). The data materials are protected at all stages of processing with the necessary physical and technical solutions. Statistics Finland has compiled detailed directions and instructions for confidential processing of the data. Employees have access only to the data essential for their duties. The premises where unit-level data are processed are not accessible to outsiders. Members of the personnel have signed a pledge of secrecy upon entering the service. Violation of data protection is punishable. 

Further information: Data protection | Statistics Finland (stat.fi) 

The data materials are protected at all stages of processing with the necessary physical and technical solutions. The unit-specific data of the data are confidential. Data are handled only by persons who need the data in their work. The use of data is restricted by usage rights.

The data are released in table formats, which does not allow for the identification of individual data producers. The time use data of groups with fewer than 50 survey days are not released in the tables.

All employees compiling the Time Use Survey have signed a pledge of secrecy, where they are obliged to keep secret the data prescribed as confidential by virtue of the Statistics Act or the Act on the Openness of Government Activities.

Statistics Finland publishes new statistical data at 8 am on weekdays in its web service. The release times of statistics are given in advance in the release calendar available in the web service. The data are public after they have been updated in the web service. 

Further information: Publication principles for statistics at Statistics Finland 

The data of the Time Use Survey are not released in identifiable format outside Statistics Finland. Statistics Finland’s Research Services provide unit-level data (excluding microdata) for scientific studies and statistical surveys only on the basis of a separate user licence decision and with identifiers removed. The microdata is disclosed to Eurostat, which prepares database tables on the data of different countries for release on its website.

Statistical data are published as database tables in the StatFin database. The database is the primary publishing site of data, and new data are updated first there. When releasing statistical data, existing database tables can be updated with new data or completely new database tables can be published.   

In addition to statistical data published in the StatFin database, a release on the key data is usually published in the web service. If the release contains data concerning several reference periods (e.g. monthly and annual data), a review bringing together these data is published in the web service. Database tables updated at the time of publication are listed both in the release and in the review. In some cases, statistical data can also be published as mere database releases in the StatFin database. No release or review is published in connection with these database releases. 

Releases and database tables are published in three languages, in Finnish, Swedish and English. The language versions of releases may have more limited content than in Finnish.   

Information about changes in the publication schedules of releases and database tables and about corrections are given as change releases in the web service. 

Revisions – i.e. improvements in the accuracy of statistical data already published – are a normal feature of statistical production and result in improved quality of statistics. The principle is that statistical data are based on the best available data and information concerning the statistical phenomenon. On the other hand, the revisions are communicated as transparently as possible in advance. Advance communication ensures that the users can prepare for the data revisions.

The reason why data in statistical releases become revised is often caused by the data becoming supplemented. Then the new, revised statistical figure is based on a wider information basis and describes the phenomenon more accurately than before.

Revisions of statistical data may also be caused by the calculation method used, such as annual benchmarking or updating of weight structures. Changes of base years and used classifications may also cause revisions to data.

The quality of the Time Use statistics is assessed at several different stages of the statistical process.

Quality management requires comprehensive guidance of activities. The quality management framework of the field of statistics is the European Statistics Code of Practice (CoP). The quality criteria of Official Statistics of Finland are also compatible with the European Statistics Code of Practice. 

Further information: Quality management | Statistics Finland (stat.fi) 

Data are released to all users at the same time. Statistical data may only be handled at Statistics Finland and information on them may be given before release only by persons involved in the production of the statistics concerned or who need the data of the statistics concerned in their own work before the data are published. 

Further information: Publication principles for statistics 

Unless otherwise separately stated in connection with the product, data or service concerned, Statistics Finland is the producer of the data and the owner of the copyright. The terms of use for statistical data.