Work ability and work participation – working-age population classification comparison results

(Text continues after graph)

Classification data about the holistic concept of work ability

In the study, we used The state of work ability in Finland data, which is a 90% population sample of the Finnish working-age population, people aged 20–64 at the end of 2021 in this case. The combination of register data sets for 2021 included a total of 2,920,099 people. The data consists of the register data of Statistics Finland, the Social Insurance Institution of Finland Kela, the Finnish Centre for Pensions and the Finnish Institute for Health and Welfare.

We aimed to study the work ability of the working-age population by operationalizing the factors proven to be related to work ability (see Lederer et al., 2014) from the register data with a particular focus on individual level dimensions, which include physical, psychological, social and demographical factors as well as factors related to economic status, education and occupation.

The register correlates for the social and demographic factors and factors related to economic status, education and occupation concerning work ability were age, gender, mother tongue (Finnish, Swedish, other), level of education (primary or missing information, secondary, higher degree), the available income, the type of municipality of residence (city, urban settlement, rural area), the size of the household (1, 2, 3 or more persons), the type of family (living alone, couples, couple with children, single parents), the number of children aged under 18 in the family (0, 1, 2–3, 4 or more) and occupational classification (10 different classes).

In addition, the register correlate for the physical and psychological dimension was the health issues index developed by us, which was scored by each period of a maximum of 30 days of receiving sickness allowance increasing the score by one. If the period of receiving sickness allowance was more than 30 days, the score increased by two. The purchase of medications under the Kela special reimbursement system increased the score by two. The use of specialized health care also increased the score by two. If there were no aforementioned sickness allowance periods or the use of services or purchase of medications, the index score was zero. As a result, the index was scored on a scale of zero to six. The indexation of health and scoring various factors highlighted various factors related to health and work ability in a multidimensional way. The score provided a multidimensional interpretation of health status for the indicator.

We carried out the classification with the means of a cluster analysis. We used the K-prototypes method in the cluster analysis (Huang, 1998), which allowed us to group an observation unit to a single group by using both continuous and categorical variables. We tested the optimal number of groups before the classification. In practice, we tested different numbers of groups by starting with two groups and ending with 20 groups. The appropriate number of groups minimizes the internal deviations from the average within groups, which makes the groups as condensed as possible. Based on the register correlates, population sample and method used in this study, the final number of groups was 14.

The results make it possible to compare groups with regard to work ability dimensions insofar as we have been able to observe the dimensions empirically. The groups can be viewed nationally or by wellbeing services county.

Groups with different levels of work ability

The interactive table below allows for viewing the groups by different register correlates or variables.

Examples of the formation of groups include groups 9 “Middle aged people with high education and a family, good health” and 10 “Middle aged people with high education and a family, poor health”, which resemble each other in many ways. For example, the gender, level of education and family type variables are very similar in these groups. However, the health status of these groups is remarkably different. This is only one way to compare the groups. A description of all groups can be found below.

Group 1 includes middle-aged people, 41 years old on average, who live in a city and a single parent family. The vast majority of them are women. About half of them have a higher education degree. They have some health problems (the health issues index 1.5 on average). The group is titled “Middle aged single parents with high or medium education living in a city, moderate health”.

Group 2 includes young people, 30 years old on average, who live alone in a city with no children. About a quarter of them have a higher education degree. They have no health problems (0.8). The group is titled “Young people with low education living alone in a city, good health”.

Group 3 includes over middle aged people, 54 years old on average, who live alone. The group includes more people who live in rural areas and urban settlements than other groups. About a fifth of them have a higher education degree and they have some health problems (4.3). The group is called “Over middle aged people with low education living alone, poor health”.

Group 4 includes young people, 30 years old on average, who live in a family with children. They typically have a low level of education. They have relatively little health problems (1.1). This group also includes relatively lot of people who live in rural areas and urban settlements. Nearly 80% of the people in this group are women. The group also includes more people who speak a language other than Finnish or Swedish as their mother tongue than any other group. The group is titled “Young people with low education and a family, good health”.

Group 5 includes young people, 29 years old on average, who live in a city with no children. A majority of them live together with another adult. About a fifth of them have a higher education degree. They have some health problems (2.7). The group is titled “Young people with low education living in a relationship in a city, poor health”.

Group 6 includes nearly retirement aged people, 61 years old on average, who live in a relationship with no children aged under 18 living in the same household. About two thirds of the people in this group are men and approximately half of the people in this group have a higher education degree. This group also includes relatively lot of people who live in rural areas and urban settlements. They have some health problems (3.3). The group is titled “Nearly retirement aged people with high or medium education living in a relationship, poor health”.

Group 7 includes over middle aged people, 59 years old on average, who live alone. A little under two thirds of them are women. About half of them have a secondary education degree. They have only little health problems (1.1). The group is titled “Over middle aged people with low education living alone, fairly good health”.

Group 8 includes young people, 29 years old on average, who live in a city in a relationship with no children. Nearly three quarters of them are men and one in four have a higher education degree. They have no health problems (0.0). The group is titled “Young people with low education living in a relationship in a city, excellent health”.

Group 9 includes middle aged people, 43 years old on average, who live in a family with children. Nearly two thirds of them are women. More than 90% of them have a higher education degree. They have very few health problems (0.5). The group is titled “Middle aged people with high education and a family, good health”.

Group 10 includes middle aged people, 41 years old on average, who live in a family with children. Nearly 80% of them are women. About 70% of them have a higher education degree. They have poor health (3.0). The group is titled “Middle aged people with high education and a family, poor health”.

Group 11 includes middle aged people, 38 years old on average, who live in a family with children. The majority of them have a secondary education degree and less than 15% of them have a higher education degree. Compared to other groups, the majority of the people in group 11 live in a rural area or an urban settlement. They have very few health problems (0.4). The group is titled “Middle aged people with low education and a family, good health”.

Group 12 includes over middle aged people, 55 years old on average, who live in a family with children. The majority of them, about 80%, are men. Nearly two thirds of them have a secondary education degree and about a fifth have a higher education degree. One third of them live in a rural area or an urban settlement. They have some health problems (1.4). The group is titled “Over middle aged people with low education with a family, moderate health”.

Group 13 includes nearly retirement aged people, 60 years old on average, who far more commonly live in a rural area or an urban settlement than other groups, are in a relationship and have no children living with them. Two thirds of them are men. More than two thirds of them have a secondary education degree. They have good health (0.5). The group is titled “Nearly retirement aged people with low education living in a relationship in a rural area, good health”.

Group 14 includes middle aged people, 51 years old on average, who live in a relationship in a city and have no children living with them. Nearly four out of five of them are women. 77% of them have a higher education degree. They have fairly good health (0.8). The group is titled “Over middle aged people with high education living in a relationship in a city, fairly good health”.

Work participation in different groups

The classification data in combination with employment data can be used to study and compare the amount of employment in groups with different factors related to work ability. The groups’ amount of work participation in 2021 can be studied by selecting the data entity “Work participation” in the interactive table below.

The average number of employment months for all groups is 8.7. There is considerable variation between the average number of employment months of different groups. The group with the most employment months is group 9, “Middle aged people with high education and a family, good health” with 11.5 months. The lowest number of employment months was in group 6, “Nearly retirement aged people with high or medium education living in a relationship, poor health”, who had 6.0 employment months. The number of unemployment months among people in this group was 0.8. This indicates that many of the people in group 6 are outside work life. This applies to most groups with young people. Young people are typically outside work life as students, on parental leave or in military service.

Groups’ work participation in different wellbeing services counties

The classification data and employment data in combination with wellbeing services county data can be used to study and compare the amount of employment in groups with different factors related to work ability in different wellbeing services counties. The interactive table below allows for making two different types of comparisons, i.e. by selecting a group and comparing wellbeing services counties or by selecting a wellbeing services county and comparing groups within this wellbeing services county.

There is significant variation between wellbeing services counties in the employment months of people in group 7, “Over middle aged people with low education living alone, fairly good health”. Within this group, the people living in Åland have the most employment months, about 7.9. People living in the North Karelia wellbeing services county have the lowest number of employment months in this group, about 5.3.

Conclusion

We offer three different ways to study the work ability of the working-age population with the classification data.

First of all, the classification results indicate what types of groups of the working-age population can be formed with regard to work ability if the population is studied by grouping on the basis of factors related to work ability. The sizes of the groups and the comparison of their characteristics offers one way of looking at the distribution of work ability factors in the Finnish working-age population. It is worth taking into account that, although we describe the groups based on the average characteristics of the people in the group, not all people in the group fully represent the average description. With regard to the family type, it should be noted that the family type of people aged over 18 living with their parents is always the same as the family type of their parents: for example, people classified as “couple with children” may not be one of the couple but a child aged over 18 living with the couple.

Additionally, the comparison of the employment months of the groups offers information about the distribution of work in the Finnish working-age population. This method highlights the fact that groups with different work ability factors have different amounts of work participation.

A third way of looking at the data sheds light on the differences between individual groups with regard to work participation in different wellbeing services counties. The observed variation indicates that people with certain work ability factors have varying degrees of participation in work life in different wellbeing services counties. Studying these observations may provide the wellbeing service county insight into how the work participation of different groups can be increased.