Mood, as a quality of feeling at a particular time or even a persistent emotion influencing our thoughts and actions, is important for personal and social adjustment. The assessment of such states in an efficient and accurate manner is a type of Holy Grail for clinicians. A solution to this quest has even greater value where there are difficulties in answering due to literacy problems or neurological conditions such as aphasia (Arruda et al., 1996a; Kontou, Thomas, & Lincoln, 2012). The resort to visual analogue scales as a self-estimate of gross mood states had its origins around 98 years ago in personnel and industrial psychology (Hayes & Patterson, 1921), but has been slow to find recognition in mainstream psychological testing even though it has been used readily in medicine or health with the assessment of pain (Haefeli & Elfering, 2006), the rehabilitation of stroke patients (Vickery, Evans, Lee, Sepehri, & Jaheen, 2009) or temperament (Yeung & Wong, 2019). Visual analogue approaches began life as graphic rating scales and the Visual Analog Mood Scales (Stern, 1997; Stern, Daneshvar, & Poon, 2010) are part of this zeitgeist. This paper reviews and evaluates the Visual Analog Mood Scales. The format and nature of the scales is reviewed together with the development, standardization and scoring. The findings of validity and reliability are also appraised and a holistic evaluation of the utility of a visual analogue assessment of mood is outlined. The purpose is to examine their utility for personal assessment, psychological interventions and mood research.

The Format and Nature of the Visual Analog Mood Scales [TOP]

The Visual Analog Mood Scales were developed originally by Stern (1997) and his co-workers for clinical use in the assessment of mood for neurological cases (e.g., aphasia, stroke) but have also found application in clinical settings as a screening tool (Bennett, Thomas, Austen, Morris, & Lincoln, 2006). There is no substantive theory of mood that underlies the Visual Analog Mood Scales other than to provide an assessment of eight clinically relevant, specific, internal mood states: afraid, confused, sad, angry, energetic tired, happy or tense. On the whole it appears to be phenomenological rather than providing a bio-medical or neuropsychological substrate for moods. The eight scales are presented each on a single page and are unipolar dimensions, such as “neutral-sad” or “neutral-afraid”. Each mood state has an ideogram or face (what we would now call an emoji) at the bottom portraying an emotion. There is an unbroken vertical line of 100mm from the bottom of the page to another ideogram or face with the word “Neutral” at the top of each page. A sample scale is illustrated in Figure 1. Six scales relate to a negative mood (afraid, confused, sad, angry, tired or tense) and two relate to a positive mood (energetic, happiness). The test taker marks the level of their mood and this is measured in millimeters from the upper edge or neutral position. The Visual Analog Mood Scales are designed for adults aged 18 years and over. Completion of the task is not demanding as it is largely non-verbal or pictorial in its presentation.

Empirical Studies Involving the Visual Analog Mood Scales [TOP]

The following section provides a summary of empirical studies (including scoring, reliability and validity) related to the Visual Analog Mood Scales. This was derived from a search of the literature in PsycINFO using the search term “Visual Analog Mood Scales”. This yielded 35 citations of which 24 provided data of some sort on the use of the scales in empirical research.

Conclusions [TOP]

As far back as 1997, Ahearn evaluated pictorial scales for the assessment of mood and concluded that they had value: “These scales are simple to complete, ensuring a high rate of compliance, and they have been shown to possess high reliability and validity. Provided that clear instruction is made to the patient, such scales can be used confidently to obtain helpful self-report information, particularly about the course and variability of illness from the patient's perspective” (Ahearn, 1997, p. 577). As a unique assessment method, the Visual Analog Moods Scales follow in this tradition. They enable (a) a description of a person’s functioning; (b) the diagnosis of problem areas; (c) to ascertain the need for therapy; and (d) even monitor the progress of an intervention on positive or negative moods. Yet, the Visual Analog Moods Scales first published just over 20 years ago have found limited application in psychological or health settings despite their potential. The reports are limited mainly to small sample sizes with clinical groups such as Parkinson’s (Foster et al., 2013), Alzheimer’s (Stern et al., 2004), stroke (Thomas et al., 2013), aphasia (Corsten et al., 2015) or dementia (Temple et al., 2004).

Although the eight scales form part of the Visual Analog Mood Scales booklet, there is no a priori justification for combining the results to produce an overall “mood” score. The conceptual nature of this overall mood is questionable. It may be reasonable, however, to discern an elevated mood level across the eight scales (with scoring reversed for the positive moods) but only as a broad indicator of the altitude of the prevailing emotional states. The adding together of eight different raw scores or even T-scores is not justified mathematically as there are no units of mood (Michell, 2005). There are also doubts about the value of the norms from the standardization group based on the comments of the review in the Mental Measurement Yearbook (Leung, 2003) and for this reason clinicians might be better advised to use the ordering of the raw scores (Cliff & Keats, 2003) with possibly 50 as an initial screening point for further clinical investigation of the negative moods.

The actual concurrent validity correlations that are cited in Table 1 do have greater meaning than one might imagine at first sight. The minimum concurrent validities for each scale (i.e., the second last row in Table 1) were related to real world criteria. Validities from meta-analyses and reviews (Meyer et al., 2001) are listed for comparison with the minimum concurrent validity correlations of the same value from the Visual Analog Moods Scales (Table 3). The comparisons are conservative and attest to the potential value of this approach to the assessment of mood, especially since the concurrent validities are not corrected for unreliability or range restriction.

A holistic evaluation of the overall utility of the Visual Analog Mood Scales might also be considered in this conclusion. In 2016, Francis, McPheeters, Noud, Penson, and Feurer developed an 18-item checklist of evaluative criteria to assess patient reported outcomes that may be used to evaluate the strengths or weaknesses of different measures. This checklist has been adapted and applied to the Visual Analog Mood Scales and the results are summarized in Table 4. The scales meet at least two-thirds of the criteria in this evaluative approach. The major failings appear on this approach to be in the areas of reliability and validity. The prevailing inference is that the Visual Analog Moods Scales have promise as a non-verbal or quasi pictorial assessment of moods but no claim is made that the Visual Analog Moods Scales offer a full or complete representation of moods.

Turning now to the psychometric foundation of the scales, it is noted that the essential foundation of each of the Visual Analog Mood Scales is that of a self-estimate based on a single measure as opposed to an inventory with multiple sets of questions relating to a mood state. For some time, there has been a view about the inherent accuracy of self-perception (Bem, 1972) because individuals can be valid observers of themselves (Duval & Wicklund, 1972). In those contexts where there is no reason for disguise or concealment, Mischel (1977, p. 248) contended that “what the person tells us directly turns out to be as valuable an index as any other more direct sign”. Writing from a perspective of self-evaluation of personality characteristics, Burisch (1984, p. 225) went further and concluded: “if self-ratings are (a) directly communicable, (b) the ultimate in economy, and (c) also more valid than their questionnaire counterparts, then we will have to face the embarrassing question of just why we continue to construct personality inventories at all”.

Aside from the issue of the validity of the Visual Analog Mood Scales as self-estimates, there is a powerful argument for their utility. They have been applied in the assessment of difficult clinical groups in which other forms or methods of assessment would struggle. Although, the results to date inspire some confidence, the data supporting the validity or reliability of the findings is limited. There is some suggestion of validity through the effect of therapeutic interventions. In terms of scoring, there is no clear evidence of the advantage of T-scores or a norm-referenced interpretation, whereas a criterion-referenced approach might have more value. Patients with major depression or mild depression (N = 56) were used to establish a cut-off point of 50 raw score for the Sad scale in the manual for the scales (Stern, 1997). As a general rule, a clinical cut-off point of scores greater than 50 for the negative moods and less than 50 for the positive moods might be tested. The combination of ratings to form a total mood score is not supported at present as there is limited theoretical foundation for an overall mood concept. The elevation of the profile may have diagnostic significance but this has yet to be explored in detail.

There is scope for a substantial program of future research to build upon the findings of this evaluation. Nine potential theoretical and practical gaps in research related to the Visual Analog Mood Scales might involve: (a) standardization and the development of representative norms for clinical samples and the general population in order to enhance the utility of the results; (b) examination of whether raw scores or a transformation of raw scores has value for clinical interpretations; (c) examination of the value of analogue ratings as self-estimates within a theory of self-perception; (d) determining the specificity and sensitivity of the results for clinical groups (e.g., chronic pain, post-traumatic stress disorder, anxiety, depression, thought disorders, neuro-cognitive difficulties); (e) locating the eight mood scales within a theory of moods (positive and negative) or within a broader theory of personality (e.g., neuroticism, extraversion, openness to experience, agreeableness or conscientiousness); (f) provision of evidence of long-term test-retest reliability of moods as some of the studies used only 15-minute time intervals; (g) validation of the results in terms of prediction and consequential validity; (h) examination of aspects of administration such as any effects of the order of presentation of the scales; or (i) whether the graphics and presentation of the figures need to match the more modern emoji ideograms is a moot point and one for separate evaluation but worthy of consideration in any future review of the scales.

In summary, it is considered that the value of the Visual Analog Mood Scales is that they provide a standardized but also efficient assessment of feeling at a particular point in time. As expected, the available evidence is that the Visual Analog Mood Scales have some reliability and accuracy. They enable inferences about the moods relevant to personal or social adjustment and it is believed that they should have found greater application in psychological theory and research.