The Social Avoidance and Distress Scale (SADS) is a 28-item true/false self-report measure that assesses two related aspects of social anxiety: subjective distress, discomfort, or negative emotions in social situations, and the tendency to avoid, of or desire to escape from social interactions.
The SADS was designed to measure behavioural and affective responses to social situations rather than cognitive aspects of social anxiety, and emerged from research recognising that individuals who are anxious in social situations may show different patterns of distress versus avoidance, and that measuring these responses would have heuristic value for understanding social anxiety (Watson & Friend, 1969).
The theoretical foundation of the SADS distinguishes it from measures of fear of negative evaluation, which focus on apprehension about others’ judgments. While these constructs are related, Watson and Friend (1969) deliberately minimised their overlap when creating the SADS to allow researchers and clinicians to differentiate between individuals who primarily fear social judgment versus those who primarily experience distress in or actively avoid social encounters.
SADS Item Content
The SADS consists of true/false items conceptually organised around two closely related domains:
Clinical Utility of the SADS
The SADS can be used to assess social anxiety, particularly its behavioural and affective components, which distinguish it from more cognitively-focused measures. Research has demonstrated that SADS scores are meaningfully related to other psychological constructs relevant to clinical formulation and treatment planning.
Clinical Characteristics Associated with High Scores
Individuals who score high on the SADS tend to have significantly lower self-esteem and perceive themselves as having less control over rewards in their lives, exhibiting a more external locus of control (Geist & Borecki, 1982). High scorers also demonstrate a lower need for achievement (Watson & Friend, 1969), lower self-confidence, reduced need for social affiliation, lower need for dominance, a higher need for deference (Geist & Hamrick, 1983).
These findings suggest that high scoring SADS individuals may struggle with social anxiety and broader self-concept difficulties. SADS scores are also associated with problematic social media use, potentially reflecting the social compensation hypothesis, whereby individuals with social difficulties may seek online interaction as an alternative to face-to-face contact (Ahmed et al., 2021). It should be noted this correlation was small, and based on a social media addiction scale measuring general problematic engagement patterns (compulsive, excessive use), and does not make a distinction specifically for self-presentation/posting behaviours versus passive consumption.
SADS Total Score
The SADS contains 28 items, each scored dichotomously (true/false), yielding a total raw score ranging from 0 to 28. Higher scores indicate greater social avoidance and distress, with approximately half the items reverse-scored to control for response bias.
SADS Severity Classification
Watson and Friend (1969) used score ranges in their research to create severity groups. These descriptors are provided below but it should be noted that these were research classifications for experimental comparisons, not validated clinical severity thresholds:
These ranges were not validated against diagnostic criteria for social anxiety disorder, as the SADS predates modern DSM conceptualisations. Rather, they provide descriptive comparisons for interpreting where an individual’s score falls relative to the original normative distribution.
SADS Percentiles
A percentile is presented that compares the respondent’s score to Watson and Friend’s (1969) non-clinical university sample (N = 205, M = 9.11, SD = 8.01). A percentile rank near 50 indicates the score is typical for this comparison sample. Percentile labels are displayed on the right side of the graph for reference.
Given the positively skewed distribution in the normative sample and its composition of university students, percentiles should be interpreted with appropriate caution when applied to clinical populations or individuals from different demographic backgrounds.
SADS Plots
On first administration, a bar graph displays the total raw score with severity range bands (Low, Moderate, High) for reference. Non-clinical percentile labels are displayed on the right-hand side of the graph.
When the assessment is administered multiple times, a longitudinal line graph is generated to track changes in the raw score over time.
SADS Validity
The SADS demonstrates good construct validity through correlations with theoretically related measures. Watson and Friend (1969) reported convergent validity with the Manifest Anxiety scale (r = .54), the Audience Sensitivity Index (r = .76), and a strong negative correlation with Affiliation (r = -.76), consistent with the construct of social avoidance. The correlation between the SADS and the Fear of Negative Evaluation scale was r = .51 (N = 205), indicating that while related, these are distinct constructs capturing different aspects of social-evaluative anxiety. Subsequent research by Geist and Borecki (1982) found significant group differences in locus of control [F(2, 137) = 8.15, p < .001] and self-esteem [F(2, 137) = 42.15, p < .001], with high SADS scorers demonstrating greater external locus of control and lower self-esteem compared to moderate and low scorers. Discriminant validity was established through the item selection procedure, which explicitly minimised covariation with social desirability (r = -.25 with the Crowne-Marlowe scale; Watson & Friend, 1969).
The SADS demonstrated excellent internal consistency (KR-20 = .94), indicating high item homogeneity. Test-retest reliability over a one-month period ranged from .68 to .79 across samples, indicating adequate temporal stability. Subsequent research has confirmed the reliability of translated versions of the SADS, with the Japanese version demonstrating Cronbach’s α coefficients exceeding .93 in experimental studies (Nomura et al., 2020).
The SADS is scored as a unidimensional measure, yielding a single total score. While Watson and Friend (1969) conceptually organised items around two related domains during scale construction, these were designed to be combined rather than scored separately. The correlation between the two subscales was .75, supporting this unidimensional scoring approach and reflecting the close interrelationship between avoidance behaviours and subjective distress in social contexts.
Normative Data for the SADS
Normative data for the SADS are derived from Watson and Friend’s (1969) original validation study, which included 205 university students (145 females and 60 males) in Canada. This non-clinical sample had a mean score of 9.11 (SD = 8.01). The distribution was positively skewed with a modal score of zero, indicating that high levels of social avoidance and distress were relatively uncommon in this sample.
Sex differences were observed in the normative sample: males scored higher (M = 11.2) than females (M = 8.24), a statistically significant difference. This pattern suggests that males in university settings may report greater social avoidance and distress than females, though these norms should be interpreted with caution when applied to other populations/settings.
Research studies have used varying score ranges to create comparison groups. Watson and Friend (1969) used different cutoffs across their validation experiments. These were research classifications for experimental comparisons, not validated clinical severity thresholds. Approximate Characterisation Based on Normative Sample. Based on the normative distribution (N=205, M=9.11, SD=8.01, mode=0):
These ranges were not validated against diagnostic criteria for social anxiety disorder, as the SADS predates modern DSM conceptualisations. Rather, they provide descriptive comparisons for interpreting where an individual’s score falls relative to the original normative distribution.
A Reliable Change Index (RCI) was calculated using the Jacobson-Truax method, based on the internal consistency of KR-20 = .94 and the standard deviation of 8.01 from the original validation sample (Watson & Friend, 1969). A change of 5 points or more on the total score represents statistically reliable change at a 95% confidence level. When monitoring treatment progress, a reduction in the SADS total score of 5 points or more indicates reliable improvement in social avoidance and distress, while an increase of 5 points or more would indicate deterioration. Given the scale’s total score range of 0–28, this RCI represents approximately 19% of the total scale range.
The SADS captures how much a person avoids social situations and experiences distress in them, making it particularly useful for quantifying the severity of social avoidance and discomfort, with features such as wanting to get away from people, avoiding unfamiliar others, and feeling tense or nervous in social situations. Higher scores on the SADS are associated with lower self-confidence, reduced desire for social connection, and greater deference to others, while very low scores are generally positive but may indicate may be associated with particularly high social confidence or assertiveness; this context can help clinicians understand how social avoidance fits within a person’s broader interpersonal style. Because the SADS measures avoidance and distress irrespective of cause, it has transdiagnostic utility and can be informative for clients whose social difficulties arise from depression, autism, trauma, or general anxiety, not only those with a primary social anxiety presentation, particularly when social withdrawal reflects distress or avoidance rather than preference or differences in social motivation.
It’s important to note that people vary naturally in their desire for social interaction and comfort with different social situations. The SADS measures distress and avoidance, but cultural background, personality traits (e.g., introversion/extraversion), sensory sensitivities, communication style differences, and personal preferences all influence social behaviour. Higher scores indicate someone is experiencing distress or avoidance, but clinical significance depends on whether this pattern is causing problems in the person’s life or conflicts with their values and goals.
No. The SADS was developed well before modern DSM conceptualisations of social anxiety disorder. It was designed as a single total scale reflecting two highly related components, social avoidance (avoiding being with, talking to, or escaping from others) and social distress (negative emotions like feeling upset, tense, or anxious in social interactions), rather than as a diagnostic or screening tool. The SADS captures core features that overlap with social anxiety disorder but does not assess other situational fears typical of the diagnosis, such as eating in front of others or scrutiny fears. Its clinical value lies in quantifying how much someone avoids social situations and how much distress they experience in them.
While lower SADS scores are generally associated with more adaptive social functioning, research suggests that very low scores (0-1) may have a nuanced interpretation. Watson and Friend (1969) found that individuals with low Fear of Negative Evaluation (FNE) scores, a related but distinct construct, exhibited higher dominance (r = -.50), suggesting that low social-evaluative anxiety may be associated with a stronger drive for interpersonal influence. For the SAD scale specifically, Geist and Hamrick (1983) found that individuals with low SAD scores showed significantly higher need for Dominance compared to those with high SAD scores, as well as lower need for Deference. This suggests that very low scores may not simply reflect comfortable social functioning but could indicate a stronger orientation toward interpersonal influence and a reduced tendency to defer to others’ judgments. The clinical relevance of any score—whether very low, moderate, or high—depends on whether the person’s social patterns are working for them, causing distress, or conflicting with their values and life goals.
Research indicates that elevated SADS scores are associated with a range of psychological difficulties beyond social anxiety itself. Geist and Borecki (1982) found that high SADS individuals tend to have lower self-esteem and perceive themselves as having less control in their lives (an external locus of control). High scorers also show less desire for social connection and a greater tendency to go along with others’ wishes.
These findings suggest that social avoidance and distress may co-occur with broader difficulties in self-concept and interpersonal functioning. Clinicians working with individuals who score high on the SADS may consider assessing these related areas as part of comprehensive treatment planning. For example, the Maladaptive Schema Scale (MSS) may be useful for further exploring these associated difficulties, particularly the Defectiveness/Shame schema (reflecting low self-esteem), Fatalistic/External Locus of Control schema (reflecting perceived lack of control over life outcomes), and Subjugation schema (reflecting the tendency to defer to others).
Watson, D., & Friend, R. (1969). Measurement of social-evaluative anxiety. Journal of Consulting and Clinical Psychology, 33(4), 448-457. https://doi.org/10.1037/h0027806
Ahmed, O., Siddiqua, S. J. N., Alam, N., & Griffiths, M. D. (2021). The mediating role of problematic social media use in the relationship between social avoidance/distress and self-esteem. Technology in Society, 64, Article 101485. https://doi.org/10.1016/j.techsoc.2020.101485
Geist, C. R., & Borecki, S. (1982). Social avoidance and distress as a predictor of perceived locus of control and level of self-esteem. Journal of Clinical Psychology, 38(3), 611-613. https://doi.org/10.1002/1097-4679(198207)38:3<611::AID-JCLP2270380320>3.0.CO;2-U
Gkika, S., & Wells, A. (2015). How to deal with negative thoughts? A preliminary comparison of detached mindfulness and thought evaluation in socially anxious individuals. Cognitive Therapy and Research, 39, 23-30. https://doi.org/10.1007/s10608-014-9637-5
Nomura, T., Kanda, T., Suzuki, T., & Yamada, S. (2020). Do people with social anxiety feel anxious about interacting with a robot? AI & Society, 35(2), 381–390. https://doi.org/10.1007/s00146-019-00889-9
Watson, D., & Friend, R. (1969). Measurement of social-evaluative anxiety. Journal of Consulting and Clinical Psychology, 33(4), 448-457. https://doi.org/10.1037/h0027806
Geist, C. R., & Hamrick, T. J. (1983). Social avoidance and distress: its relationship to self-confidence, and needs for affiliation, change, dominance, and deference. Journal of clinical psychology, 39(5), 727–730. https://doi.org/10.1002/1097-4679(198207)38:3%3C611::AID-JCLP2270380325%3E3.0.CO;2-H
