# Some thinking about measurement in psychology

In this section we will look at the different ways there are to perform measurements on people. Two stand out:

- Self-report, where people answer questions about properties or ideas of themselves or of thoughts they have considering a certain topic. Methods of this kind are also called ‘explicit methods’ as participants in a study are explicitly asked to formulate an answer, tick a box in a survey, or otherwise perform a prescribed behaviour needed to answer the question posed.
- Implicit measurement, where people do not know a measurement is taken or do not know what the study they participate in is about.

The presentation below draws heavily on Dijksterhuis (2011) and presents a brief historical overview of a methodological look at research methods in psychology. Ever since the dawn of scientific and experimental psychology its students and scholars have been thinking about the special problems that measurement in this field possess. A complete overview is impossible to give, but I’ve made a selection of specific ideas, from a range of psychological sciences, that one may want to take into account when performing behavioural research. Fasten your seat-belts, next follows a dense introduction of a lot of psychology.

Up front, I believe that the below constitute good reasons to, in psychological measurement, often favour implicit measurements over explicit measurements. There are exceptions, that I will indicate. One difficulty is that the term ‘implicit’ in this context is not used to mean the exact same thing by different scientists in the field. De Houwer and Moors (2007) seem to equate it to ‘automatic’, referring to the type of psychological _process_ that results in an observable outcome (like a behaviour) of a psychological measurement (see also De Houwer 2006). Others (Köster 2003, Dijksterhuis 2011, Dijksterhuis et al. 2022) rather refer to the measurement _circumstances_ (hence calling the measurement _method_ ‘implicit’) than to the type of psychological process.

There is much more to say about the distinction between ‘implicit’ and ‘explicit’ methods. This will take us outside the realms of this introductory text, so we have to leave that be for now.

## William James’ snare of psychology

Where else to start than with William James (1842-1910), who talks about ‘snares of psychology’ in his book ‘Elements of Psychology’, in the chapter called “The methods and snares of psychology” (James 1890) James writes about 'the named state' which should be distinguished from 'the naming state'. This means that a (self-)report of a certain psychological state, say sadness, is made while the subject is in another state. The act of reporting such a state (e.g. saying: “I feel sad today.”) is not the same thing as the reported state itself (the sadness felt). In fact some reflection is needed in order to be able to report sadness. One has to interpret his/her bodily and psychological circumstances to be able to, after the fact, report something like 'I am sad', or 'I currently feel sadness’. The reporting is a certain state in itself, which may have exerted an influence on the memory of the previous state which one tries to report. James' conclusion is that the introspection and report of one’s own internal psychological states is a snare.

One can conclude from this that introspection is an invalid measurement method. It also has another implication for psychological measurement. A researcher who receives information directly from his/her subject in an experiment cannot be certain that this information is complete, accurate or useful. If the researcher is interested in subjective verbal utterances of subjects in a certain experiment, this is of course a good way of obtaining those. If the researcher wants to use the information for other means, e.g. to predict future behaviour of the person or group under investigation he/she cannot automatically assume that the subject has given complete and accurate information. This is why psychological experiments can be very complicated and why often the subject is rather _not_ to know what exactly is going on. Any cues available to the subject can interfere with the information that he/she provides the researcher with. Strictly controlled experiments enable the researcher to conclude which information is useful, as outcome of the experiment, and which was an effect of cues that were –sometimes inadvertently- given to the subject. In psychological measurement this means that anything our experimental subjects tell us researchers cannot be taken as gospel. We must always take the circumstances under which it was reported into account. This is a great responsibility of psychological scientists. We must see the report in the context of what we know about the psychology of our subjects. In some cases we may already suspect upfront what they can and cannot report and how they will perceive certain situations.

## Reasons are not causes

The Dutch psychologist Johannes Linschoten (1924-1964) in his book ‘The psychologists fallacies’ (Linschoten 1964), explains the difference between a ‘motive’ and a ‘motivation’. The difference between these two terms is subtle, and may be peculiar for the Dutch language Linschoten wrote in. In Dutch 'motivation' ('motivatie' in Dutch) is the complex of factors that stimulates and aims behaviour, while 'motive' ('motief' in Dutch) is a rationalisation, justification or pretext given, e.g. for a certain behaviour. This illustrates that the latter (‘motief’) is the result of a state that reflects upon another state. The former (‘motivatie’) can be an underlying cause of behaviour, not necessarily the same as the ‘motief’, and not necessarily known to or accessible for the subject itself.

Another, similar, idea figures in the concept of 'illusory ownership' of ones deeds (Wegner 2002). Wegner argues that it is possible to be mistaken about the sources of one’s own action. Ones perception of conscious will in those cases is illusory. This can mean that a choice one makes is not always made because one consciously wanted it, but it can be _perceived_ to be our own choice only after the fact. The theory of apparent mental causation proposed by Wegner and Wheatley (1999) calls into question the common sense view that conscious will is the cause of behaviour. It accounts for misattributions to one’s own will by suggesting that when people think about an action in advance of its occurrence, and alternative sources of the action are not known, people will feel they are the source of action.

So behaviour is not always the result of consciously willing it. Wilson (2001) writes about the 'adaptive unconscious', shaped by evolution, guiding us through a complicated environment, and helping us with decisions through intuition. This all happens without our explicit knowledge, and often even without the possibility of knowledge. The processes responsible are inaccessible to conscious awareness. Or, as Wilson (2004) has put it: ‘People can no more observe how they are unconsciously categorizing their environments, setting goals, and generating intuitions than they can observe how their kidneys work.’.

## Emotions and feelings

Damasio (2001) explains the difference between an emotion and a feeling, much like James (1890) has introduced this distinction. The former is the bodily reaction to a stimulus, and the latter the conscious interpretation by the subject of that emotion. As William James put it in 1890, 'It is: “I cry therefore I am sad.”, but not “I am sad therefore I cry.”'. This may sound like a topsy-turvy situation, but it makes sense when one realizes that the same bodily reaction, crying (an emotion), can be a sign of sadness as well as of joy, depending on the situation. This means that the emotion is attributed a _reason_, but this does not mean that this attribution is the very _cause_ of the emotion; reasons are not necessarily causes.

The distinction between emotion and feeling is important, especially now in recent years the word ‘emotion’ is on the lips of many applied psychological (marketing and consumer) scientists. Sometimes the word is used without any background knowledge of the phenomenon, showing an inability to differentiate between the everyday use of the word and its scientific meaning. When a subject _reports_ a specific emotion, it has thereby become conscious, a feeling. This also means that emotions can go unnoticed, but can have an effect on behaviour nevertheless. This may be contrary to the everyday use of the word ‘emotion’, which rather is more like the ‘feeling’ as explained above.

## Irrational decision making

The underlying causes may provide a non-rational route to decision making, alongside the reasons resulting from a rational route. In many economical decision making contexts Kahneman and Tversky (1979, 2000) have amply illustrated that many people make decisions that can be shown to be sub-optimal from a strict rational, or utilitarian, viewpoint (cf. Kahneman et al. 1982). As a result subjects may actually gain less money in comparison with the alternative they do not choose. Damasio (2003) suggests that this is the result of an 'emotional route' to decision making which may supersede the rational route.

Both psychologists Daniel Kahneman and Gerd Gigerenzer (Gigerenzer et al. 1999, Gigerenzer 2023) study human behaviour and the choices people make. Superficially their work is very comparable and may lead to the same conclusion. However, when comparing the two approaches a difference shows. Gigerenzer et al. suggest that the non-rational (or ‘intuitive’) decisions are adaptive. They are heuristics that help guiding behaviour, also when the circumstances do not allow for rational pondering. This may occur when there is not enough time or when the information is incomplete. Kahneman et al. rather see the non-rational decisions as errors in the human rational decision making system.

## Preferences and inferences

Another valuable line of research can be summarised by ‘Preferences need no inferences’, which is the title of a 1980 paper by the Polish-born American psychologist Robert Zajonc (1923-2008). In 1984 two papers were published in the 'American psychologist', one advocating the primacy of cognition (Lazarus 1984) and the other the primacy of affect (Zajonc 1984). These two papers illustrate two opposing viewpoints about human choice and motivation. Lazarus (1984) argues that it is necessary first to analyse a situation, and to build some knowledge about it, in order for an affective response to occur, an approach known under the name of appraisal theory. Zajonc (1984) argues the opposite; emotion/affect is primary, only after that may knowledge about a situation occur, but cognition is not a prerequisite for affective responses to occur. A respectable number of studies show that emotion has primacy over cognition (cf. Bechara et al. 1997, 2000, Kihlstrom 1999). In one telling experiment (see Zajonc 1980) subjects were tachistoscopically presented with words, flashed too short to enable identification. However, this presentation allowed an appropriate affective response to the words. The affective value of the stimulus is somehow registered but the identity of the stimulus remained unknown to the subjects.

The conclusion from Zajonc’s work should be that, not only need one not think about preferences, one cannot know the real reason (‘motivation’) of one’s own choices, due to the independence of the affective and cognitive systems. In an evolutionary perspective this makes perfect sense. All animals, humans included, in critical situations do not have time to cognitively analyse the situation nor to ponder alternative strategies. The ones that took their time are either eaten by predators or died from starvation.