1 50 Years of Social Indicators

The article by Land and Michalos provides an excellent overview of developments in the social indicators movement over the last 50 years and the present state of play. It makes clear what a huge amount has happened and been achieved. Notwithstanding the fact that the 2009 report by Stiglitz et al. (2009) devoted little attention to these developments (cf. Noll 2011), recent initiatives such as the OECD’s Better Life Initiative (OECD 2016) show that lessons have been learned from the social indicators movement and that the findings serve as a basis for building further. Land and Michalos conclude their review of the last 50 years with an agenda for the future. I believe this is an excellent agenda to take up and take forward.

Yet one is left with a nagging feeling when reading the article by Land and Michalos. Surely we can only move forward once we have settled the debates of the last 50 years and reached broad agreement on what exactly has been achieved? Land and Michalos’ article describes many developments, many different initiatives and a number of key discussion points. I will attempt to distil those discussion points from their article and to arrange them in a number of principles for measuring quality of life. It may then be possible to view the four pointers for follow-up research described by Land and Michalos at the end of their article, as a coherent whole.

In doing this, I will not limit myself to Land and Michalos’ article, but will take it together with a number of other overview studies, such as those by Hagerty et al. 2001 and Noll and Berger (2014). Based on these studies, I will formulate seven items which could form the starting point for further research on quality of life (see also Boelhouwer 2016). The principles are based on what has been done and achieved over the last 50 years:

2 Focus on People

The focus when studying quality of life is on people. This implies that the primary object of the research is not policy, system characteristics or institutions, for example—though these aspects can of course be important when interpreting differences and developments.

3 More than Economics

Quality of life research involves adopting a perspective that goes beyond economics alone. However, the importance of social aspects does not mean that economic aspects do not matter: an insight into factors such as income and poverty is also relevant from the broad perspective of quality of life. However tempting it may sometimes be, especially in the light of the discussions that followed the publication of Piketty’s bestseller Capital, disadvantage and deprivation cannot be explained purely in economic terms—in the same way that it would be impossible to explain differences between groups exclusively on the basis of education level, for example.

4 Objective and Subjective

Land and Michalos rightly highlight the important debate within the social indicators movement on the question of whether quality of life should be measured using objective or subjective indicators. That debate can now be regarded as closed: there is a fairly high degree of consensus that both are needed in order to measure quality of life fully. In each of the dimensions of quality of life, it is important to separate a person’s actual situation from their perception of that situation (e.g. an objectively good home need not mean that the person concerned enjoys living there). This approach also produces insights that are of interest for policy, especially when the outcomes of the two types of indicators differ. It is important to keep the objective and subjective indicators separate from each other and not to combine them in a composite indicator or in an index.

5 Multiple Dimensions

Quality of life research involves research in several different dimensions. It is however important that the number of dimensions does not become too large, because otherwise there is a danger of losing the overall picture. The question of precisely which dimensions should be included under quality of life is often the subject of discussion. Despite this, the overview of different initiatives does allow us to derive a basic list (Table 1; also compare the meta-study by Hagerty et al. (2001)Footnote 1 and the eight domains proposed by Stiglitz et al. (2009).Footnote 2

Table 1 Domains included in a selection of Western and Central European social monitors.
Full size table

In itself, this insight is not new. There was a good deal of consensus as long ago as the 1970s regarding the choice of domains in different countries: “I was very intrigued by the fact that “my” list was very similar to the lists developed in other countries, even if the political systems and cultures were very different. […] I think that the lists also reveal a high degree of universalism in what is considered as social concerns in all countries.” (Johansson 2002, pp. 25–26).

Instead of continually arguing about the precise nomenclature, and instead of constantly developing yet another new domain set, my suggestion is to start from an existing set and to go with the domains currently used by the OECD in the Better Life Initiative: housing; income; jobs; community; education; environment; civic engagement; health; life satisfaction; safety; work-life balance. For analytical reasons and to increase their usability for policy, however, I believe these domains should not be added together (as the OECD does). I will explain this further later in this article based on a conceptual scheme that can be used to analyse and describe quality of life.

6 Outcome Versus Determinants

Quality of life is a broad concept that is concerned with the overall effect of policy. It is thus a concept that is used to measure outcomes; for example, it does not so much measure the amount of money spent on public housing or the number of homes built, but rather the quality of the homes in which people live. Interpreting quality of life as an outcome in this way also enables a (policy-relevant) distinction to be made between determinants, life opportunities or risk factors on the one hand and outcomes on the other. The determinants then comprise personal characteristics such as age and household composition, and resources such as income, education, labour market position and health. Someone with access to a large number of resources will be in a better position to achieve a good quality of life than someone with few resources.

7 Distribution and Difference

In itself, quality of life is a concept with a neutral meaning. It is only when qualifiers such as ‘good’, ‘poor’, ‘more’ or ‘less’ are added that it acquires a normative character. Adding such descriptors is however necessary in order to be able to make statements about the increase or decrease in quality of life: are things going better or worse? An insight into the distribution of quality of life, the differences between groups in society, provides policy information on accumulation of disadvantage and, if there is a perceived need, provides pointers for policy changes or adaptations.

Ideally, data at a personal level are needed to obtain an insight into distribution aspects. Only when information is available on all relevant domains for the same person is it possible to say anything about accumulation effects.

8 An Index?

One major area of debate is whether indicators from different domains should be combined to form a single index. Among the arguments in favour are that an index makes clear at a glance whether things are going better or worse (just as GDP does for the economy). An index is easy to communicate and simplifies comparisons between groups. On the other hand, an index is a simplification of developments in the different domains; there is no uncontroversial method for weighting the indicators; and it is reasonable to ask whether we are not comparing apples and oranges.

I would place a question mark after this last point, which has been the subject of considerable debate over the last 50 years. Whilst I believe we will reach consensus, I also believe that this is not essential: a ‘dashboard’ consisting of a limited set of indicators and an index which summarises those indicators need not be mutually exclusive. In my opinion, there is little wrong with some people using a dashboard and others compressing that dashboard to a single index. Hence my question mark.

Yet it would be a good thing if consensus could be reached on the need for an index to measure quality of life. The reason for this was given as long ago as the 1970s by the economist Drewnowski (1974): only combining social indicators into a single figure will deliver an equivalent alternative to set against GDP—itself also a combined indicator. The GDP index was also not established overnight or without discussion; it is by no means obvious to everyone what it comprises and its meaning is the subject of debate. These are the same arguments that are often levelled against a quality of life index, but they do not alter the fact that GDP is a widely used measure.

When it comes to weighting the different components—another point for debate with regard to indices—there are sufficient good solutions available (see also the OECD Handbook on constructing composite indicators 2008). This is moreover a topic that will always remain the subject of debate. The trick now is to accept this and then choose a method anyway: ‘good enough’ is better than striving for perfection or consensus. In the absence of clear guidelines, opting to apply equal weights to the different domains would appear to be a pragmatic but defensible choice. Let’s make a start and see where it leads!

My preference for an index does not mean that such an index would have to include everything. As also suggested by Stiglitz et al. (2009), it is perfectly possible to encapsulate quality of life in an index, but it is not sensible to include sustainability within that index, for example. To speak in terms of sustainability: quality of life here and now can be encapsulated in a single index, but quality of life elsewhere or at a later time is something that must be considered separately.

9 A Conceptual Framework for the Land–Michalos Agenda for the Future

As far as I am concerned, we should move quickly towards finding consensus on the topics mentioned above. We can then take forward the recommendations and agenda for the future as set out by Land and Michalos. It is helpful to see those recommendations collectively and to start from their fourth recommendation: theory building. When discussing the issues arising from social indicators research over the last 50 years, I briefly mentioned that it could be relevant for both analysis and policy to separate resources from outcomes and subjective from objective indicators. Resources allow people to shape their lives based on their own choices and on a range of norms and values (comparable with items 1 and 2 on Land and Michalos’ agenda. Governments play an important role in providing those resources and ensuring that they are distributed in a way that is regarded as equitable.

This theory building can be represented schematically in a conceptual framework, as shown in Fig. 1 (I have discussed this framework at greater length elsewhere; see Boelhouwer 2010). Figure 1 is based on the Dutch measurement of quality of life using the SCP Life Situation Index, but this could easily be substituted by a measurement of quality of life such as I discussed earlier, based on the domains in the OECD Better Life Initiative.

Fig. 1
figure 1

Taken from Boelhouwer (2010, p. 166)

Conceptual framework for describing quality of life.

Full size image

How satisfied and happy people are depends in part on their objective circumstances, but by no means entirely. It is therefore better to keep the two measures of quality of life separate but to include them both in the analysis. This can generate policy-relevant insights into the processes which lead to one person being happy with their situation and another person not.

10 Conclusion

The article by Land and Michalos provides an excellent description of what has been achieved over the last 50 years in social indicators and quality of life research. The question now is whether we can move forward with that description. My own view is that it would be useful first to identify and establish the things on which we are agreed, so that it is clear where the many different initiatives are genuinely building on earlier results. And in reality, we also need to move towards greater synergy and unity. The question for me is whether we will now move forward in unity or divided. My vote is for unity. That involves reaching consensus on a number of crucial issues, such as which dimensions make up quality of life, distinguishing between determinants and outcomes, or analysing objective and subjective indicators separately. But the maxim ‘forward in unity’ also implies reaching a consensus on the need for an index. Only then can social indicators play the same role in the political debate as the economic indicators encapsulated in GDP do in the economic debate.

The social indicators and quality of life research arena is almost bursting at the seams with all the different initiatives. Bringing some order and calm to this situation requires an overarching and authoritative body to play a coordinating and steering role. As far as I am concerned, we could from now on work with the domains proposed by the OECD.