Affective phenomena (emotions, mood and affect) have a long research and theorizing history. The issues are complex and subject to controversy. Therefore, there is an overwhelming amount of literature present, which one should dig into deeply before being able to understand and contribute to this subject. This paper focuses on measurement in sport and exercise psychology.

Which three steps should one consider when choosing a measure?

The first step you should take when measuring something in the domain of affective phenomena, is choosing which construct you are going to target: emotion, mood or core affect. Secondly, you should choose which of many theoretical models you’re going to adhere to. Only then you evaluate the psychometric information (whether a measure meets the criteria for reliability and validity).

What are the differences between affect, emotion and mood?

• Core affect is a neurophysiological state consciously accessible as a simple primitive non-reflective feeling. Core affect is constantly experienced, although the nature and intensity of affect may vary over time. It can be a component of emotions and moods, but can also occur solemnly. Examples are pleasure and displeasure, tension and relaxation.

• Emotions are a complex set of interrelated sub-events concerned with a specific object. People often show behavior that fits with the emotion they are experiencing. Emotions are elicited by something, reactions to something and are generally about something. Therefore, the cognitive appraisal between the person and the object causing the emotion is seen as a defining element.

• Moods lasts longer than emotions and there are not as specific. Also, the cause of a mood may not always be easy to identify.

It is important to keep these differences in mind when choosing a measure for your research because these different concepts should be measured in different ways.

You also have two different approaches: the distinct-states approach and the dimensional approach. According to the distinct-states approach, each state is a distinct entity. It therefore focuses on the unique features of the different states. However, some researchers say that the states are not completely independent. They propose underlying dimensions and came up with the dimensional approach. There are two accepted dimensional models:

1. The circumplex model (Russell, 1980) which is based on the idea that the affective states can be placed in a model with two axes: the affective valence and the perceived activation.

2. The two-dimensional solution (Zevon & Tellegen, 1982; Watson & Tellegen, 1985) proposes two axes: high-activation pleasant affect and low-activation unpleasant affect.

These models are variants of one another, not fundamentally different. However, the last model caused much confusion in the literature because of the naming of the dimensions. Therefore, Thayer (1989) proposed a third variant with the axes energetic arousal (energy-tiredness) and tense arousal (tension-calmness). These models all explain plenty of variance in the core affect dimension (which it is limited to).

Watson and Tellegen (1985) understood the limitations of dimensional models, and therefore proposed that the affective domain should be seen as having a hierarchical structure. Broad dimensions can be captured by these models, but the distinct-states approach is required to examine uniqueness. We can conclude from this that both the distinct-states and dimensional approach can be used, and which one is more suitable depends on the aim of the study.

What specific measures can be used?

To say something about affective responses, an increasing number of studies employ psychophysiological measures. We will briefly mention the different categories and measures that are often used.

Single-item dimensional measures of affect

Single-item measures are short to take. They therefore minimize the impact it has on participants. On the other hand, since it only measures one response (which could be erroneous) they tend to be less reliable than multi-item measures. Some famous single-item tests are:

• The Self-Assessment Manikin (SAM; Bradley & Lang, 1994; Lang, 1980)

• The Affect Grid (AG; Russell, Weiss, & Mendelsohn, 1989) based on Russell’s circumplex model of affect.

• Feeling Scale (FS; Hardy & Rejeski, 1989)

• The Felt Arousal Scale (FAS; Svebak & Murgatroyd, 1985)

Multi-item measures of distinct mood states

Multi-item measures take longer to administer than single-item measures. As a consequence, people can get tired from taking the test and may cause reactivity to the testing. However, they are less susceptible to error. Two multi-item tests are:

• Multiple Affect Adjective Checklist (MAACL; Zuckerman & Lubin, 1965)

• Profile of Mood States (POMS; McNair et al., 1971)

Multi-item dimensional measures of affect

Dimensional measures were developed in order to capture a global domain, and therefore measure multiple dimensions that are supposed to underlie this domain.

One example of a multi-item dimensional measure of affect is the Positive and Negative Affect Schedule (PANAS; Watson et aI., 1988). However, this test has two big limitations: 1) the items represent a mixture of emotions, moods and affects and 2) there seems to be an inconsistency between the conceptual model that formed the basis and the actual content and structure of the test.

Another dimensional test is the Activation Deactivation Adjective Check list (AD ACL; Thayer, 1989) which measures energetic arousal and tense arousal. Also, a measure that has a similar structure as the AD ACL is recently developed and is called the Four-Dimension Mood Scale (4DMS; Gregg & Shepherd, 2009). This last test may be more sensitive to the effects of physical exercise.

Multi-item measures of specific emotions

Anxiety is the emotion that received the most attention in exercise psychology. The most frequent test used to measure this emotion is the State-Trait Anxiety Inventory (STAI; Spielberger, Gorsuch, & Lushene, 1970; Spielberger, 1983). This test measures both state anxiety (characterized by subjective and consciously experienced thoughts and feelings) and trait anxiety (seen as a personality trait). Despite its popularity, the STAI received some criticism as well.

Multi-item measures of specific moods

In exercise psychology, research on specific moods has focused on depression. Measures like the Beck Depression Inventory (BDI; Beck, Steer, & Garbin, 1988) and the Hamilton Rating Scale for Depression (HRSD; Hamilton, 1960) are used most of the time.

Exercise-specific measures of affect

Researchers of exercise psychology found that because of the unique stimulus that exercise is, it should be measured in a different way than by the domain-general measures of affect. Therefore, the Exercise-Induced Feeling Inventory (EPI; Gauvin & Rejeski, 1993), the Subjective Exercise Experiences Scale (SEES; McAuley & Courneya, 1994), and the Physical Activity Affect Scale (PAAS; Lox, Jackson, Tuholski, Wasley, & Treasure, 2000) were developed. However, these measures lack an underlying theory and the items are based on a non-representative sample.

What should researchers and practitioners pay attention to?

As mentioned earlier, the study of mood, emotion and affect is not simple. A very long research history, literature collection and measurement methods are available. The challenge for any researcher in this field is to completely understand all possibilities and make an informed choice for their own research. One should at all times follow the three-step plan provided in this chapter and the rationale for each decision should be outlined in your research paper. For a complete overview of all mentioned tests in this chapter, you can take a look at table 28.1 in which all self-reporting measures are summarized.