(666, Brie Instruments for Treatment Plansing, Monitoring, and Outcome Assessment worked on in therapy. Ay of the next 21/ areas (A.'T. Beck et al,, 1996, p.'5) cant be assumed an area of difficulty if it receives a score of 3: 1, Sadness 2. Pessimism 3, Past failure 4, Loss of pleasure 5. Guilty feelings 6, Punishment feelings 7. Self-dislike 8, Solf-criticalness 9. Suicidal thoughts or wishes ing 11. Agitation 12, Loss of interest 13, Indecisiveness 14, Worthlessness 15, Loss of energy 16, Changes in sleeping pattern 17, Ievitability 18, Changes in appetite 19. Concentration difficulty 20, Tiredness or fatigue 21, Loss of interest in sex by strong endorsements (2 or 3) on items 9 (suicidal thoughts or wishes) and 2 (pes-_ ‘simismm), Whereas th level of depression (based on total score) and presence of specific item endorsement can assist in suggesting the presence of a formal DSM-5 disorder.) STATE TRAIT ANXIETY INVENTORY treatment planning and to establish the impact of interventions. The State-Trait Anx- iety Inventory (STAI: Spielberger, Gorsuch, Lushene, Vagg, & Jacobs, 1983) is ideally suited for this purpose because it is a brief (40-item), self-report inventory that is easy to understand (with a sixth-grade reading level) and is sensitive to transitory episodes of anxiety (states) as well as more stable personality features that predispose a client to experiencing more chronic levels of anxiety (traits). Respondents are asked to respond on a 4-point scale (“Almost Never” to “Almost Always” forthe trait scale and “Not atState Tralt Amaety Inventory 667 AII” to “Very Much So” for the state scale) to items related to their current anxiety and their general state of worry. The STAI is currently the most frequently used measure ‘of anxiety with over 8,000 studies availuble inthe literature, Research has evaluated its use in relation to phobias, test anxiety, pani, generalized anxiety, and the impact of specific types of treatment, such as cognitive behavior therapy, systematic desen- sitization, relaxation, and rational emotive therapy (Spielberger, Sydeman, Owen, & Marsh, 1999). It has also been used extensively in cross-cultural research and has been translated into more than 60 languages and dialects. Construction of the STAT began in 1964 with a single st of tems that could be used toassess ther state or trait anxiety based on rewording the instructions (Form A). The state instructions requested clients to complete items for how they felt “right now, at ‘the moment,” whereas the trait descriptions asked them to indicate how they generally feel. The items were originally derived and adapted from existing anxiety inventories, including the Affect Adjective Checklist (see Spielberger & Reheiser, 2004). Items were reduced and the scale was refined based on the degree to which individual items corre- lated with the Manifest Anxicty Scale, Anxiety Scale Questionnaire, and Welsh Anxiety Scale of the MMPI (Spielberger et al, 1999). Further evaluation with Form A indi- cated that merely rewording the instructions was not sufficient to eliminate the ce trait connotations of some of the items. For example, the item “I worry too mud was a good measure of trait anxiety, but merely rewording the instructions indicated that it was not a good measure of state anxiety. As a result, a second form (Form X; Spielberger et al., 1983: also see Spiclberger & Reheiser, 2004) was developed based on the trait and state dimensions having their own individual items, Trait items were selected based on their having the highest correlations with the Manifest Anxiety Scale, Anxiety Scale Questionnaire, and Welsh Anxiety Scale, as well as being the most sta-_ ble over time, The state items were selected based on their being most sensitive to high- versus low-stress conditions (high construct validity) and having the highest internal consistency, ‘A docade after the publication of Form X, the STAT underwent a further major revision based on factor analysis, a clearer understanding of the concept of anxiety and an attempt to eliminate item overlap with depression, This resulted in the current (Form Y) version having 20 items for trait and 20 for state anxiety (Spielberger & Sydeman, 1994), Form Y was normed on 1,838 employees of the Federal Aviation ‘Administration, 855 university students, 424 high school students 1,701 Air Force recruits, and 263 naval recruits. Older persons and those with more education scored somewhat lower than those who were either younger or less educated, which suggests ‘it might be important to use age and education-related norms, Additional norms are available for a neuropsychiatric population, general medical/surgical patients, and young prison inmates. ‘Children (STAIC; Spielberger, 1973), is also available. Reliability and Validity Test-retest reliability for college students over 30- and 60-day intervals indicated rea- “sonably good coeflcients ranging between 73 and 86 for trait anxiety. In contrast state anxiety test-retest reliabilities were relatively lower, ranging from .51 for males to(668 Bret Instruments for Treatment Planing, Monitoring, and Outcome Assessment 36 for females (Spielberger et al., 1983). ‘The lower range for state anxiety is expected, given that state anxiety is considered a more changeable construct, susceptible to many {transient factors, Given the expected fluctuations for state anxiety, measures of internal consistency would be more appropriate and important to consider, These have resulted in quite high state anxiety median coeflicients ranging between .88 and.93, with a sim= larly high median trait anxiety coefficient ranging from .92 to .94 (Kabacoff, Segal, Hersen, & Van Hasselt, 1997: Spielberger et al, 1983). ‘The content validity of the STAI trait scale is supported in that five out of a possible ‘eight domains for a DSM-1V-based diagnosis of a Generalized Anxiety Disorder were reflected in the items (Okun, Stein, Bauman, & Silver, 1996). Concurrent validity is supported by high correlations with the Manifest Anxiety Scale and the Anxiety Seale Questionnaire, which ranged from .73 to .75 (Spielberger & Reheiscr, 2004). These cor relations are sufficiently high that the STAI can be considered an alternative measure of trait anxiety. However, the STAT has the advantage of being shorter and less contami- inated by measures of depression, Lower and moderate, but still significant, correlations wore found between the STAI trait scale and the Worry Seale (.57) and Padua Inventory (57: Stanley, Beck, & Zebb, 1996). The construct validity of the STAI is suggested in that psychiatric patients gener- ally have higher scores on trait anxiety than nonpatient groups (Spielberger etal, 1983: Stanley ct al. 1996). One exception is that, as expected, pationts with personality dis- orders tended to have lower scores (Spielberger et al., 1983). Kabacoff etal. (1997) also found that patients with anxiety disorders had slightly higher STAI trait scores than patients without anxiety disorders, Despite this support for the convergent and diver- gent validity of the STAI, Kabacoffet al. were not successful in developing adequate ‘cutoff scores for identifying the presence of an anxiety disorder, This was primarily because of difficulty finding a score that produced both good sensitivity (high identifi- cation of true positives) and good specificity (high identification of true negatives) Construct validity for the validity of the STAI state anxicty scale is supported in that students during in-class exams and military recruits undergoing stressful training pro- cedures had higher scores when compared to scores taken after relaxation procedures or with age-matched controls (Spielberger et al., 1983). Numerous studies have demon- strated that scores on the STAL are sensitive to the impact of a wide variety of inter ventions (Newham, Westwood, Aplin, & Wittkowski, 2012; Spielberger et al., 1983; Speilberger & Reheiser, 2004), The STAT has had difficulty discriminating patients with anxiety from patients with depression (Kennedy, Schwab, Morris, & Beldia, 2001), though because of the high comorbidity and general distress related to both syndromes, ‘most measures of anxiety face this same problem. Factor analyses of the STAI have been mixed. According to the STAI scale devel- ‘opment expectations, there should ideally be one factor that loads on trait anxiety and ‘another one on state anxiety (Spielberger et al., 1983). In contrast, Bieling, Antony, ‘and Swinson (1998) found a higher-order factor derived from the trait anxiety items they referred to as negative affect and two lower-order factors that they concluded were organized around depression and anxiety. Thus, the trait items seemed not to be pure measures of anxicty but included measures of negative affect and depression in ‘addition to anxiety, Whereas Spielberger and Sydeman (1994) did attempt to make Form Y more of a pure measure of anxiety than Form X, this seems to have beenState Tralt Amxety Inventory 669 ‘only partly successful. The difficulty in developing a pure measure of anxiety underlies the issue, frequently found in other measures of anxiety and depression, that anxiety _and depression have overlapping features with correlations typically ranging between |AS and .75 (Lovibund & Lovibund, 1995). The factor structure of the STAI is further complicated in that Kabacoff et al. (1997) found two factors related to whether the items were worded in a positive or negative direction. They concluded that these items “were “method factors” unrelated to the constructs of anxiety. An alternative model by Vigneau and Cormier (2008) found a four-factor model based on a combination of state and trait, as well as whether the items were worded as these variables being present ‘orabsent (State Anxiety positive items, State Anxiety negative items, Trait Anxiety neg- ative items, Trait Anxioty positive items). In other words, the bipolar aspects (wording ‘of the items) of trait/state anxiety emerged as independent factors. Use with Diverse Groups ‘The utility of the STAT has been demonstrated in a number of cross-cultural contexts. For example, Hishinuma et al. (2000) evaluated the STAI with an Asian Pacific ado- lescent population and found good internal consistency and a factor structure that ‘was similar to the normative group. Similarly, a Spanish translation of the STAI has been found to have good internal consistency and a comparable factor structure, sug- ‘gesting it can be used with Spanish-speaking populations (Novy, Smith, Rogers, & Rowzee, 1995). Overall, the STAI has excellent adaptability across a wide variety of ethnic and cross-cultural groups (Spielberger, Moscoso, & Brunner, 2004; Vigneau & ‘Cormier, 2008). The STAT has also been found to have good psychometric properties with older adult populations (Bergua et al., 2012; Kabacoff, Segal, Hersen, & Van Hasselt, 1997: K. Kvaal, Ulstein, Nordhus, & Engedal, 2005). It has been normed for adolescents and thas been found to have good psychometric properties with this population (Spielberger. 1983), Interpretation Because the STAI includes two unidimensional subscales, interpretation is mainly €om= ‘posed of descriptions for the variables being measured. Interpretation should consider both the variable itself and the relative magnitude of the person's score.(610, Brie Instruments for Treatment Plansing, Monitoring, and Outcome Assessment RECOMMENDED READING Antony, M. M., & Barlow, D. (Eds). (2011). Handbook of assessment, treatment planning, and ‘outcome evaluation: Empirically supported strategies for psychologial disorders (nd ed.) New York, NY: Guilford Press. Groth-Marnat, G- (1999), Financial efficacy of clinical assessment: Rational guidelines and issues for future research, Journal of Clinical Psychology, 55, 813-824 Maruish, M. E, (2004). The use of psychological testing for treatment planning and outcomes assessment rd ed.). Mahwah, NJ: Erfbaum, Maruish, M. E., & Nelson, E. A. (2014). Psychological testing inthe age of managed behavioral ‘health care. Mahwah, NJ: Enbausn,