'Generalised' or 'particularised' happiness measures? implications for time diary research into happiness

The Importance of Happiness The study of happiness – and well-being more generally – has achieved ever-greater salience within the social sciences over the last two decades. Much effort has gone into producing reliable measures of happiness within individuals and populations. This has been driven by a desire to produce credible alternatives to more conventional economic measures of societal progress. Objective economic indicators like GDP-growth and per-capita income have traditionally been used as stand-alone indicators of national social progress, and their popularity has stemmed from the ease with which such measures of material and physical well-being can be closely monitored. It has been assumed that these measures alone will adequately reflect well-being. However, recent work on measuring happiness, which is gaining traction in the eyes of the public and policymakers, suggests this is not the case. Interest in measuring and incorporating happiness into policy has been fuelled by well-publicised analyses of people’s self-reported levels of happiness from population surveys. Such surveys ask generalised questions about happiness, along the lines of: “Taking all things together, how happy would you say you are, 00 being extremely unhappy and 10 being extremely happy?” Studies aggregating these measures to the level of population contrast such measures with rates of GDP growth. They have found that if a country’s level of GDP is high, it does not immediately follow that levels of subjective well-being (such as feelings of happiness or satisfaction) will match this (Easterlin, 1974, Layard, 2005, Diener and Seligman, 2004, Kahneman, 2006, Veenhoven, 1994). Subjective indicators like happiness tell an important story. They are, at the very least, important measures to include alongside the more objective conventional economic indicators in national accounts, if we are to have a proper and well-recognised understanding of all the factors that comprise human well-being and progress. The task of generating appropriate measures of happiness, however, is not straightforward. Popular research by Diener and Seligman and Layard has advanced both the profile of happiness and the measurement of happiness by use of the generalised happiness question, but we should not conflate the appropriateness of the former with the latter. We suggest that there are particular contentions to be raised over the use of the generalised happiness measure. Generalised happiness versus particularised happiness The primary criticism we raise against the generalised happiness measure is that survey questions that simply ask respondents how happy they feel in a general ‘all of life’ sense cannot capture the day-to-day micro level factors that affect a person’s feeling of happiness, such as social interaction with others. While emotions like happiness are commonly perceived to be individual, private phenomena, sociologists of emotion have also, in the last thirty years, shed light on their social dimensions. Some sociologists have argued that cultural norms, otherwise known as ‘feeling rules’ are central to an individual’s emotional experience. These rules can determine what is felt and when, and how intensely it is felt. Others say that emotions are also shaped by more social structural or macro-level factors. Thus, this paper will explore and discuss some of the major sociological studies of emotion (Hochschild, 1979, 1983, 1998, Durkheim, 1961, Thoits, 1990, Shott, 1979, Kemper, 1981) and will shed light on some of the socio-cultural factors that can influence people’s feelings, and more importantly, their perceptions of their levels of happiness. However, what these researchers have in common is an understanding that emotions like happiness are contextual – one way or another, they are caused by events and activities that people experience. This suggests that people’s responses to survey questions on generalised happiness are likely missing a great deal of important and subtle information, and must be examined with an eye towards looking for viable alternatives. Contrasting ‘generalised’ and ‘particularised’ happiness using the ESS We will present data from the European Social Survey (ESS) to illustrate the nature of the relationship between generalised measures of happiness and alternative measures of happiness and well-being. Alternative measures are particularised in that they are asked in relation to the past two weeks, and the activities and experiences that occurred therein. We will demonstrate that these measures do not show the kind of statistical coherence that one would expect if they were adequately examining the same phenomenon, and argue that this is representative of the inadequacy of the generalised happiness measure, and of the need for alternative measures. Alternative particularised measures of happiness – Emotion Time Diaries We then proceed to argue that a better method for measuring happiness is through the use of time diaries. This builds on an existing body of research. Mihaly Csikszentmihalhyi (2003) did a study using the Experience Sampling Method (ESM) and Daniel Kahneman (2004) developed the Day Reconstruction Method (DRM) to measure happiness and other emotions. Both go into great depth, looking on a micro level at everyday activities and how these affect emotions and happiness. The former asks respondents to record their feelings in response to random beeps delivered at regular intervals throughout the day, whilst the latter uses a diarylike structure and intends to act as a less expensive alternative to ESM. For two days, at the end of each day, respondents must retrospectively record activities that they engaged in, as well as emotions felt at the time. However, whilst either method captures the multi-dimensional nature of happiness (such as the different types of activities that could make one feel happy), they both have the distinct disadvantage of only being able to uncover emotions that were brought about by an activity or event at the present moment, and not at any time in the past. There is also no way of knowing whether the activity itself was the cause of the emotion, or whether it was caused by something different (for example, an event that occurred in the past, or something other than an activity). Neither method provides a means by which socio-cultural influences on emotion (explored by sociologists in the last thirty or so years) can be detected. We propose an alternative methodology – an Emotion Time Diary - for quantitatively measuring happiness and other emotions. This aims to rectify all of the problems that have been encountered in studies using both survey measures and the Experience Sampling and Day Reconstruction Methods. It takes the diary-like structure from these two methods but it also has its differences. The design will enable the accounting for socio-cultural influences on emotion that are discussed in the sociological literature. We will present aspects of the overall design of the diary and some preliminary results from a small-scale pilot exercise. This will shed light on the advantages that the diary method of happiness measurement has over conventional generalised survey variables.