Skip to main content
PMC home page
International Journal of Methods in Psychiatric Research logoLink to International Journal of Methods in Psychiatric Research
. 2018 Nov 8;28(1):e1751. doi: 10.1002/mpr.1751

Evidence for the reliability and preliminary validity of the Adult ADHD Self‐Report Scale v1.1 (ASRS v1.1) Screener in an adolescent community sample

Jennifer Greif Green 1,✉, Gerrit DeYoung 1, Mary Ellen Wogan 1, Erika J Wolf 2, Kathleen Lynne Lane 3, Lenard A Adler 4
PMCID: PMC6877133  PMID: 30407687

Abstract

Objectives

There is a need for brief and publicly‐available assessments of attention deficit hyperactivity disorder (ADHD) easily administered in large‐scale survey efforts monitoring symptoms among adolescents. The ADHD Self‐Report Scale v1.1 (ASRS; Kessler et al., 2005) Screener, a six‐item measure of ADHD symptoms, is a valid and reliable screening instrument for ADHD among adults. The current study provides initial evidence for the reliability and validity of the ASRS Screener among a community sample of U.S. adolescents.

Methods

Middle and high school students in grades 6 through 12 (N = 2,472) completed the ASRS Screener, along with the Strengths and Difficulties Questionnaire (SDQ; Goodman, 2001) and several questions about school functioning.

Results

The ASRS Screener demonstrated good internal consistency, with items captured by a single underlying latent variable, which was invariant across subsamples differing by gender. The ASRS Screener scores were associated with the SDQ subscale measuring hyperactivity/inattention (r = 0.58) and significantly less strongly associated with other SDQ subscale scores (r = −0.15–0.41). The ASRS Screener scores were also significantly associated with student‐reported school functioning.

Conclusion

Findings suggest directions for future research and provide preliminary support for use of the ASRS Screener as a brief tool for identifying symptoms of ADHD among adolescents.

Keywords: adolescents, attention deficit hyperactivity disorder, schools, screener

1. INTRODUCTION

Attention deficit hyperactivity disorder (ADHD) is a common child/adolescent disorder, with an estimated U.S. national lifetime prevalence of 8.7% (Merikangas et al., 2010) and 12‐month prevalence at 6.5% among adolescents (Kessler et al., 2012). ADHD has also been implicated as a particularly powerful predictor of poor educational outcomes (e.g., reduced academic achievement and increased school absences and tardiness) in studies controlling for comorbid internalizing and externalizing disorders (Breslau et al., 2009; Breslau, Miller, Chung, & Schweitzer, 2011; Kent et al., 2011; McLeod, Uemura, & Rohrman, 2012). As such, schools monitoring the psychological needs of students can benefit from data on population‐level ADHD prevalence and the capacity to examine its correlates and consequences. In particular, a number of surveillance systems are used by schools to monitor youth health and mental health needs (e.g., the Youth Risk Behavioral Surveillance System; California Health Kids Survey), but these do not include questions about ADHD symptoms. Given the negative outcomes associated with ADHD, there is a need for brief and efficient, publicly‐available measures that can be used to monitor ADHD symptoms in youth.

One challenge in monitoring symptoms of ADHD among youth is that there are few brief self‐report measures designed for adolescents. Although Diagnostic and Statistical Manual (DSM)‐IV and DSM‐5 criteria require a pattern of inattention and/or hyperactivity/impulsivity symptoms (American Psychiatric Association, 1994, 2013), a number of studies have demonstrated that hyperactivity/impulsivity symptoms become less observable with age, whereas deficits in attention and executive functioning become more noticeable (Biederman, Mick, & Faraone, 2000; Hart, Lahey, Loeber, Applegate, & Frick, 1995; Wolraich et al., 2005). Executive functioning (EF) refers to cognitive processes that allow people to regulate their thoughts and behaviors, including exhibiting attentional control and cognitive inhibition. For example, EF skills include the ability to prioritize complex tasks, sustain focus on challenging intellectual activities, and attend to fine details. Research with adults has indicated that assessments specifically focused on EF deficits may better predict clinical ADHD diagnoses than assessments measuring symptoms of inattention, hyperactivity, and impulsivity (Kessler et al., 2010). To more effectively assess ADHD in adulthood, a World Health Organization (WHO) workgroup, in collaboration with the WHO World Mental Health (WMH) Survey Initiative, developed the Adult ADHD Self‐Report Scale v1.1 (ASRS; Kessler et al., 2005). Originally an 18‐item symptom checklist, the scale was reduced to a brief six‐item screener, reflecting the DSM‐IV diagnostic criteria for ADHD. The measure has demonstrated strong reliability and validity in adult community and clinical samples (Adler et al., 2006; Fuller‐Killgore, Burlison, & Dwyer, 2012; Kessler et al., 2007), as well as a non‐ADHD sample of adults (Silverstein, Alperin, Faraone, Kessler, & Adler, 2017). Further, prior research found evidence that the ASRS showed measurement invariance in the form of equal factor loadings across subpopulations of adults based on gender and age (Morin, Tran, & Caci, 2016).

Although originally developed for adults, the ASRS 18‐item Symptom Checklist and six‐item Screener have increasingly been used in studies of adolescents. The advantages of the ASRS for use with adolescents are that it is a publicly‐available and brief assessment of ADHD that has been translated into multiple languages and is worded such that it might more readily reflect adolescent experiences (e.g., deficits in EF) than measures designed for children. Several studies have demonstrated the validity of the ASRS in comparison to clinician‐administered interviews in adolescent clinical populations. In one, Adler et al. (2012) administered the ASRS Symptom Checklist to a sample of 88 outpatients ages 13–17 years old who met criteria for DSM‐IV ADHD. They found ASRS Symptom Checklist scores demonstrated high internal consistency (α = 0.93–0.94) and significant correlations with the clinician‐administered ADHD rating scale (Faries, Yalcin, Harder, & Heiligenstein, 2001; r = 0.72–0.73). In another, Sonnby et al. (2015) administered the ASRS v1.1 Screener to a sample of 134 outpatients between the ages of 12–17 years ‐old in Sweden. They found good internal consistency (α = 0.79) and a strong correlation between the ASRS Screener and the clinician‐administered Kiddie Schedule for Affective Disorders and Schizophrenia (Kaufman et al., 1997) ADHD symptom severity scale (r = 0.51). The ASRS Symptom Checklist has also been used in a number of other studies of adolescent mental health, including a large population‐based study of 16–19 year‐old adolescents in Western Norway (Sivertsen, Harvey, Pallesen, & Hysing, 2015; Sivertsen, Skogen, Jakobsen, & Hysing, 2015; Skogen et al., 2014) and in studies of social anxiety among U.S. adolescents (Keeley et al., 2018; Thomas, Daruwala, Goepel, & De Los Reyes, 2012). Yet, the psychometric properties of the ASRS Symptom Checklist and Screener have not previously been reported in a large community sample of U.S. adolescents.

The current study evaluates the reliability, factor structure, measurement invariance (across gender and school level), and preliminary validity of the six‐item ASRS Screener in a community school‐based sample of U.S. adolescents. We address the following questions: (a) Does the ASRS Screener for ADHD symptoms demonstrate adequate internal consistency and measurement invariance across gender and school level? (b) Does the ASRS Screener demonstrate convergent and divergent validity in comparison with an established measure of child emotional and behavioral symptoms across gender and school level? (c) Are ASRS Screener scores associated with student reports of school absences and tardiness?

2. METHOD

2.1. Participants and settings

Two school districts in a suburb in the northeastern United States administered anonymous assessments of mental health and school functioning as part of district‐wide efforts to improve mental health supports for students. The two districts included a total of seven elementary schools and each had one middle and one high school. A letter was sent home to parents describing the assessment and all students in grades 4 through 12 were invited to participate. Only students in grades six and above completed the ASRS v1.1 and are included in the current study. All participants completed surveys using online survey software during the regular school day. Teachers read a script describing the survey and informed students that they could choose whether to participate and that they could skip any questions that they preferred not to answer. Students were also informed that they could discuss any concerns with their teacher, school counselor, or administrator. Secondary analysis of anonymous data was approved by the Boston University Institutional Review Board.

The current study includes middle and high school students in grades 6 to 12 (ages 11–18 years old) who completed the ASRS v1.1 Screener. A total of 2,472 students completed assessments (n = 1,111 District A, n = 1,361 District B; overall response rate = 94.4%). Students identified as 52.3% female, 46.6% male, 0.7% other, and 0.3% missing (Table 1). State data indicated that 91–94% of students in the two districts identified as non‐Latino White; therefore, surveys did not include questions about race/ethnicity to protect student anonymity.

Table 1.

Sample demographics and descriptive statistics (N = 2,472)

% (n)
Gender
Male 46.6 (1,152)
Female 52.3 (1,294)
Grade
6 16.7 (412)
7 16.4 (406)
8 16.5 (409)
9 13.1 (324)
10 13.0 (322)
11 13.1 (325)
12 11.1 (274)
Never %(n) Rarely %(n) Sometimes %(n) Often %(n) Very often %(n)
ASRS
Avoid/delay effortful tasks 17.3 (419) 31.8 (771) 30.0 (727) 13.4 (324) 7.5 (182)
Problems remembering 30.1 (726) 35.4 (856) 24.1 (582) 7.3 (177) 3.1 (74)
Problems with organization 36.2 (876) 31.1 (753) 23.0 (557) 6.3 (152) 3.5 (84)
Problems finishing projects 38.6 (934) 31.4 (760) 21.4 (517) 6.0 (146) 2.6 (62)
Feeling overly active 36.8 (884) 23.7 (570) 23.7 (569) 10.3 (248) 5.6 (134)
Fidgeted/squirmed 35.8 (866) 20.9 (505) 20.2 (490) 11.9 (288) 11.2 (272)
Never %(n) Once/twice %(n) 2–3 Times a month %(n) Once a week %(n) Several times a week %(n)
Absences/tardiness
Late for class 71.1 (771) 21.2 (230) 4.9 (53) 1.0 (11) 1.8 (20)
Missed a class 66.9 (727) 23.9 (260) 5.5 (60) 2.1 (23) 1.5 (16)
Missed a day 65.9 (717) 25.6 (278) 7.1 (77) 0.9 (10) 0.6 (6)
Open in a new tab

2.2. Measures

2.2.1. Adult ADHD self‐report scale v1.1 (ASRS) screener

The six‐item ASRS Screener (Kessler et al., 2005; https://www.hcp.med.harvard.edu/ncs/asrs.php) asks respondents to rate on a five‐point Likert‐type scale (never to very often) the extent to which they exhibited ADHD symptoms in the previous 6 months (e.g., “How often do you have trouble wrapping up the final details of a project, once the challenging parts have been done?”). We calculated a total sum score for the ASRS and also used the scoring approach recommended by the ASRS v1.1 Screener manual (Adler, Kessler, & Spencer, 2003), whereby respondents who endorsed at least four out of six items were considered at “elevated” risk for ADHD. As described above, the ASRS Symptom Checklist and Screener have previously demonstrated good psychometric properties in studies with adults (Adler et al., 2006; Kessler et al., 2007) and with clinical adolescent samples (Adler et al., 2012; Sonnby et al., 2015).

2.2.2. Strengths and difficulties questionnaire (SDQ)

The Strengths and Difficulties Questionnaire (SDQ) (Goodman, 2001) is a 25‐item measure of student social–emotional strengths and challenges producing five subscales: hyperactivity/inattention, emotional problems (i.e., anxiety and depression), conduct problems, peer problems, and prosocial behaviors. Prior studies documented the reliability of the SDQ and validity in comparison with structured diagnostic interviews (Goodman, 2001; Goodman, Ford, Simmons, Gatward, & Meltzer, 2000). The SDQ hyperactivity/inattention scale includes questions that align with DSM‐IV ADHD criteria. It has been used in other research as an indicator of probable ADHD (Cuffe, Moore, & McKeown, 2009) and in one study was more strongly associated with a clinician ADHD diagnosis than the Child Behavior Checklist (Goodman & Scott, 1999). Internal consistency in our sample for the five SDQ subscales was as follows: hyperactivity/inattention α = 0.76, emotional problems α = 0.75, conduct problems α = 0.64, peer problems α = 0.63, prosocial behaviors α = 0.69.

2.2.3. School absences and tardiness

Participants were asked three questions about the frequency with which they (a) were late for school, (b) missed a class, or (c) missed a day of school in the past 12 months because of a personal or emotional problem. Participants rated their responses on a five‐point Likert‐type scale (never to several times a week).

2.2.4. Additional measures

Finally, participants provided demographic information including their grade‐level (which we used to categorize students into middle and high school levels) and gender.

2.3. Analysis procedures

To assess the ASRS Screener reliability, we calculated internal consistency for the total sample and in subsamples stratified by gender and grade level using Cronbach's alpha (α). In addition, due to concerns about assumptions underlying Cronbach's alpha and its misestimation of reliability (Sijtsma, 2009), we also calculated two alternative estimates of reliability: coefficient H and omega total coefficient (ω), as described by McNeish (2017). These estimates have the advantage of not requiring tau equivalence (e.g., items can vary in their relation to the construct measured).

We next tested the ASRS factor structure and whether it was invariant across gender (male and female) and school level (middle and high). We first fit an overall one‐factor confirmatory factor analysis (CFA) to examine the fit of the hypothesized one‐factor model. This analysis employed the robust maximum likelihood estimator (MLR) and adjusted the standard errors to account for nonindependence of observations nested within schools. After establishing the fit of the model in the sample overall, we then fit two sets of multigroup models to examine the potential for measurement noninvariance by gender and school level. In both sets of models, we first fit the CFA in each group separately (e.g., males vs. females) and then tested a series of increasingly restrictive nested models to assess for equal form (i.e., configural invariance), equal factor loadings (i.e., metric invariance), equal indicator intercepts (i.e., strong invariance), and equal error variances (i.e., strict invariance) per the procedures described by Brown (2006). Nested models were compared with the less restrictive models using the nested Chi‐square test (and adjusted for the use of the MLR estimator). When Chi‐square values for the nested model do not differ significantly from the less restrictive model with more freely estimated parameters, this suggests the more restricted model is preferred (e.g., with more parameters held invariant) because it provides equivalent model fit to the parent model, but does so with greater parsimony. All models were stratified by school and were tested using Mplus 8 (Muthén & Muthén, 1998‐2017) and evaluated using standard fit indices and criteria (Hu & Bentler, 1999). Though we tested for all forms of measurement invariance, the most important tests were those evaluating equal form and equal factor loadings. Evidence of invariance in factor loadings across groups would raise concerns about potential bias, as this would indicate the relationship between the items and the latent construct vary by group.

Second, to determine ASRS convergent and divergent validity, we calculated correlation coefficients comparing the ASRS Screener sum score to each subscale of the SDQ in the total sample and in subsamples stratified by gender and grade. Here, we expected the ASRS score to be more strongly associated with the hyperactivity/inattention subscale of the SDQ than other subscale scores. We used Williams' test to determine whether differences in the magnitude of correlation coefficients were statistically significant (Howell, 2010; Steiger, 1980). Finally, to evaluate ASRS concurrent validity, we estimated several multiple regression models, which tested the association of school absences and tardiness with the ASRS sum score, gender, and school level (middle vs. high), controlling for school district.

3. RESULTS

The average ASRS Screener score was 7.5 (SD = 4.9) and 14.6% of the sample had an “elevated” score. Scores were distributed with skewness of 0.52 (SE = 0.05) and kurtosis of −0.07 (SE = 0.10), suggesting no significant deviation from a normal distribution. Males reported slightly higher ASRS Screener scores (M = 7.71, SD = 4.84) than females (M = 7.31, SD = 4.82), though this difference was not significant in an ANCOVA controlling for grade (p = 0.050). The ASRS Screener scores increased significantly with age (from a mean of 5.4 among 6th graders to a high of 9.6 among 11th graders; F(6, 2,425) = 41.0, p < 0.001). Similarly, whereas 9.6% of middle school students reported elevated ASRS Screener scores, 19.6% of high school students reported elevated scores. The most commonly reported symptom was avoiding or delaying tasks requiring a lot of thought (Table 1) and this symptom increased with age, having an average score of 0.9 among 6th graders and 2.2 among 12th graders.

The ASRS Screener demonstrated good internal consistency in the total sample (α = 0.80; coefficient H = 0.97; ω = 0.96). Reliability was also good in subsamples stratified by gender and grade, including among the youngest students (6th grade α = .79; coefficient H = 0.98; ω = 0.95; Table 2).

Table 2.

Reliability and scores of ASRS screener in the total sample and by gender/grade

ASRS sum mean (SD) ASRS elevated score % ASRS α ASRS coefficient H ASRS ω
Total 7.5 (4.9) 14.6 0.80 0.97 0.96
Gender
Male 7.7 (4.8) 15.0 0.80 0.98 0.97
Female 7.3 (4.8) 13.7 0.80 0.90 0.89
Grade
6 5.4 (4.4) 6.2 0.79 0.98 0.95
7 6.4 (4.5) 9.5 0.77 0.98 0.95
8 7.0 (4.8) 13.1 0.81 0.92 0.92
9 7.7 (4.4) 12.5 0.76 0.99 0.97
10 8.7 (4.8) 17.1 0.79 0.99 0.99
11 9.6 (5.1) 26.0 0.80 0.99 0.98
12 9.4 (4.8) 23.5 0.75 1.00 0.99
Open in a new tab

Note. ASRS: Adult self‐report scale.

Note. α = Cronbach's alpha, ω = omega. coefficient H and ω were calculated as described by McNeish (2017).

The one‐factor CFA model evaluated in the full sample fit the data well (Table 3) and all items loaded significantly on the latent variable in the β = 0.49–0.75 range (all p < 0.001). The model also fit the data well for both males and females separately and for both middle and high school students separately. Evaluation of measurement invariance as a function of gender indicated the model structure was invariant across males and females and the relationship between the indicators and the latent variable (i.e., the factor loadings) was also invariant across males and females (Table 3). However, imposing the equal intercepts constraint on the model yielded a significant change in chi‐square, suggesting that males and females differed in observed scores on the indicators. Evaluation of measurement invariance as a function of middle versus high school‐aged students provided support for equal form, but the next most restrictive model suggested that the relationship between the items and the factors differed by middle versus high school students such that imposing the equality constraint decreased model fit (Table 3).

Table 3.

Fit of confirmatory factor analyses and test of measurement invariance

Model χ2 df RMSEA SRMR CFI TLI BIC Model comp. Δ χ2 Δ df
1. Overall 92.48*** 9 0.06 0.03 0.97 0.95 41110
2. Multi‐group: Gender
2a. Males 42.98*** 9 0.06 0.03 0.97 0.96 19086
2b. Females 67.09*** 9 0.07 0.03 0.96 0.94 21531
2c. Equal form 109.52*** 18 0.07 0.03 0.97 0.95 40642
2d. Equal factor loadings 116.24*** 23 0.06 0.03 0.97 0.96 40607 2c vs. 2d 3.91 5
2e. Equal indicator intercepts 195.61*** 29 0.07 0.04 0.94 0.94 40650 2d vs. 2e 88.99*** 6
2f. Equal error variances 207.58*** 35 0.06 0.05 0.94 0.95 40619 2e vs. 2f 12.69* 6
3. Multi‐group: School type
3a. Middle School 34.03*** 9 0.05 0.02 0.98 0.97 19,883
3b. High School 72.98*** 9 0.08 0.04 0.96 0.93 20,891
3c. Equal form 105.12*** 18 0.06 0.03 0.97 0.95 40,799
3d. Equal factor loadings 123.69*** 23 0.06 0.04 0.96 0.95 40,780 3c vs. 3d 17.67** 5
3e. Equal indicator intercepts 451.45*** 29 0.11 0.12 0.85 0.84 41,128 3d vs. 3e 369.22*** 6
3f. Equal error variances 477.75*** 35 0.10 0.11 0.84 0.86 41,116 3e vs. 3f 27.62*** 6
Open in a new tab

Note. All models used the robust maximum likelihood estimator and were stratified by school to adjust standard errors for nonindependence of observations within a school. Chi‐square difference tests adjust for use of robust maximum likelihood estimator. df = degrees of freedom; RMSEA = root mean square error of approximation; SRMR = standardized root mean square residual’ CFI = confirmatory fit index; TLI = Tucker‐Lewis index; BIC = Bayesian information criterion; comp = comparison.

*

p < 0.05.

**

p < 0.01.

***

p < 0.001.

To evaluate the source of this metric noninvariance, we examined the pattern of factor loadings in the equal form model and found that item six (fidgeted or squirmed with hands or feet when sitting for a long time) had weaker loadings in the high‐school group. Based on this finding, we tested for measurement invariance again without this item. Without item six, the equal factor loadings model provided equivalent fit to the equal form model (Δ χ2 = 6.32, Δ df = 4, p = 0.18), suggesting invariance of the factor loadings across the middle and high school groups. As with the measurement invariance testing by gender, the next most restrictive model (equal intercepts) substantially decreased model fit. In total, this provided support for metric invariance across gender and across school level, except for item six, which differed in its relationship to the latent construct for middle and high school students. When this item was removed from the ASRS screener, the rate of elevated ASRS scores among high school students decreased from 19.6 to 14.8%.

Correlations between the six‐item ASRS Screener and SDQ subscale scores are presented in Table 4 for the total sample and in subsamples stratified by gender and grade. As expected, the ASRS Screener scores were significantly more strongly associated with the SDQ subscale measuring hyperactivity/inattention (r = 0.58) than all other subscales, which measured emotional problems (r = 0.41, Williams t = 8.86, p < 0.001), conduct problems (r = 0.35, Williams t = 13.00, p < 0.001), peer problems (r = 0.30, Williams t = 13.61, p < 0.001), and prosocial behaviors (r = −0.15, Williams t = 26.62, p < 0.001). This pattern persisted in subsamples by gender and grade level.

Table 4.

Correlations between ASRS screener and SDQ in the total sample and by gender/grade

ASRS scores SDQ Inattention/hyperactivity SDQ emotional SDQ conduct SDQ peer SDQ prosocial
Total sample 0.577** 0.407** 0.352** 0.300** −0.146**
Gender
Male 0.557** 0.359** 0.321** 0.255** −0.103**
Female 0.593** 0.467** 0.389** 0.324** −0.182**
Grade
6 0.632** 0.410** 0.409** 0.284** −0.231**
7 0.638** 0.433** 0.273** 0.295** −0.164**
8 0.607** 0.457** 0.306** 0.375** −0.119*
9 0.534** 0.331** 0.393** 0.299** −0.111*
10 0.550** 0.352** 0.302** 0.150** −0.004
11 0.552** 0.276** 0.287** 0.192** −0.025
12 0.473** 0.374** 0.311** 0.263** −0.003
Open in a new tab

Note. ASRS: Adult Self‐Report Scale; SDQ: Strengths and Difficulties Questionnaire.

*

p < 0.05.

**

p < 0.01.

Finally, a series of multiple regression models estimated the association between ASRS scores and school absenteeism/tardiness, controlling for gender, school level (middle vs. high), and school district (Table 5). Results indicate that ASRS Screener scores were significantly associated with increased reports of arriving to school late, missing class, and missing an entire day of school. Associations between ASRS screener scores and these three indicators of school functioning remained significant in models that added controls for the four non‐ADHD‐related SDQ subscale scores (emotional problems, conduct problems, peer problems, and prosocial behaviors).

Table 5.

Association between ASRS screener and school absences/tardiness in the total sample

Absences/tardiness Late for school Missed a class Missed a day
B SE B β B SE B β B SE B Β
Gender (ref = male) 0.188 0.048 0.118** 0.273 0.048 0.171** 0.251 0.043 0.176**
School level (ref = middle) −0.062 0.049 −0.038 0.069 0.049 0.042 0.116 0.044 0.080**
ASRS sum 0.017 0.005 0.103** 0.019 0.005 0.114** 0.019 0.005 0.127**
R 2 0.038** 0.063** 0.071**
Open in a new tab

Note. ASRS: adult self‐report scale.

Note. All analyses control for school district

**

p < 0.01.

4. DISCUSSION

Many ADHD measures rely on parent or teacher report (Conners, Sitarenios, Parker, & Epstein, 1998; Wolraich et al., 2003), however, there are several reasons researchers and school staff might want to collect data directly from adolescents. First, many school districts are involved in school‐based surveillance of student health and mental health (e.g., the Youth Risk Behavior Surveillance System) that allows them to easily identify the prevalence of mental health needs, monitor trends over time, and identify correlates of health and mental health risks. These surveillance systems are often administered anonymously and directly to students, as they aim to determine population trends, rather than identify individual students for intervention (Dowdy, Ritchey, & Kamphaus, 2010). Second, collecting data from adolescents directly is appealing in settings (including schools) where it can sometimes be difficult to contact parents, and adolescents might therefore be the most accessible source of information on symptoms. Finally, easily observable hyperactivity and impulsivity symptoms tend to recede as students grow older, whereas inattention and EF problems remain (Kessler et al., 2010). Teachers and parents may have more difficulty identifying problems with inattention and EF than problems with hyperactivity/impulsivity, making it less likely that older students will receive necessary treatment without the use of validated screening tools.

The current study provides support for the ASRS Screener as a reliable assessment of ADHD among middle and high school students. Internal consistency was high across all three of the reliability estimates calculated, for males and females, and across all grade‐levels of respondents. These coefficients are comparable with other well‐known and longer assessments of ADHD symptoms relying on teacher and parent report, such as the revised Conners Teacher Rating (CTRS‐R; Conners et al., 1998), with Cronbach's alpha of 0.73–0.95, and somewhat lower than the Vanderbilt ADHD Parent and Teacher Rating Scales (Wolraich et al., 2003), with Cronbach's alpha of 0.90–0.95. The coefficients are also slightly lower than reports of Cronbach's alpha from the 18‐item ASRS Symptom Checklist administered to clinical and community samples of adolescents (i.e., 0.89–0.93; Adler et al., 2012; Sivertsen, Harvey, et al., 2015, Sivertsen, Skogen, et al., 2015), but higher than in adult samples completing the six‐item ASRS Screener (i.e., 0.63–0.72; Kessler et al., 2007). This pattern is expected given that longer scales generally produce higher reliability coefficients (Tavakol & Dennick, 2011).

Our analysis of measurement invariance verified that items load together onto a single factor across gender and school level. Our analyses also provided strong support for the metric invariance of ASRS items across subgroups of males and females, a result consistent with findings from a study with adults (Morin et al., 2016). However, findings from the analysis of measurement invariance by school level suggested items are differentially related to the ASRS depending on whether students are in middle or high school. Further inspection indicated that the item asking about fidgeting/squirming was more weakly associated with the ASRS construct among high school than middle school students. This is consistent with research that has found symptoms of hyperactivity decrease in late adolescence and young adulthood (Wolraich et al., 2005), and suggests that this question may be less effective for screening for ADHD among older students. The other five items demonstrated measurement invariance by school level.

Results also provide initial support for the convergent and divergent validity of the ASRS Screener in comparison with the SDQ hyperactivity/inattention subscale. The correlation between these scales (r = 0.58) is similar to, or higher than, what has been found in other studies assessing the convergent validity of ADHD symptoms and related constructs (Duckworth & Kern, 2011; McCandless & Laughlin, 2007). Further, as expected, and as found in other studies (Keeley et al., 2018), the ASRS showed significantly weaker associations with other indices of emotional and behavioral well being, suggesting divergent validity relative to other common childhood psychological symptoms. Additional studies using diagnostic gold standards are still needed to fully evaluate the validity of the ASRS. Our findings also indicate that scores on the ASRS Screener are associated with self‐reported school absences and tardiness. In models controlling for gender, school level, district, and SDQ subscale scores, the ASRS Screener is significantly associated with being late to class and/or missing a class or an entire day of school due to an emotional problem. Although we would not expect ADHD to be the main predictor of tardiness and absences and, indeed, see that these associations are only modest, this finding suggests that the same students identified by the ASRS may also struggle with functional outcomes in school.

The total percent of students (14.6%) with elevated scores is comparable in the current sample to previously published studies using the ASRS. In particular, using the ASRS Screener, Kessler et al. (2007) estimated that 14% of an adult population had elevated scores. Although our rates are higher than the estimated prevalence of a 12‐month diagnosis of ADHD among adolescents (6.5%; Kessler et al., 2012), the ASRS screener is not intended to provide a diagnosis of ADHD, but rather to screen participants who require further evaluation (Caci, Morin, & Tran, 2014). However, in the current study, the rate of elevated scores in late adolescence is particularly high (23–26%). These scores are higher than found previously in studies with adults, but lower than ASRS scores reported in other studies including among a nonclinical sample of 14–15 year‐olds (Keeley et al., 2018) and a treatment‐seeking sample of 12–17 year ‐olds who did not have a diagnosis of ADHD (Sonnby et al., 2015).

The increase in ASRS Screener scores from 6th to 12th grade is largely driven by the high percent of older adolescents who reported they avoided or delayed starting task that required a lot of thought. One possibility is that over the course of high school, students are increasingly tasked with unstructured assignments, while they are simultaneously required to plan and organize their time with more limited adult support. The slight decrease in ASRS Screener reports from 11th to 12th grade might reflect student acquisition of skills to plan and complete assignments. A second possibility is that the relatively high ASRS Screener reports among late adolescents may reflect characteristics of our sample. Both districts participating in this study had state test passage rates substantially higher than state or national averages (approximately 90%). It is possible that in this high achieving setting, students might perceive that they are not effective or efficient in their academic work when compared with their peers. Third, it is possible comorbid disorders explain the increase in screener prevalence rates in later adolescents. High ASRS scores here and in other studies with adolescents (Keeley et al., 2018; Sonnby et al., 2015), coupled with results from the analysis of measurement invariance, suggest a need for further research on the choice and weighting of ASRS items for older adolescents.

Until results are replicated, we encourage findings be interpreted cautiously with attention to the following limitations. First, our research relies on self‐report, without parent or teacher interviews or objective measures of school absences and tardiness (e.g., school records). Second, the SDQ is not a diagnostic tool and therefore results should be considered as a complement to prior studies that compared the ASRS to clinician‐administered measures in clinical samples (Adler et al., 2012). Although conducting similar clinician‐administered interviews in a large community sample would be ideal, it would also be costly and the response rate is unlikely to be as high as that obtained here. Third, the current study is limited to two high achieving school districts in the northeast with limited racial and ethnic diversity. Finally, in 2017, Ustun et al. (2017) published a new six‐item version of the ASRS Screener updated for DSM‐5 criteria, yet, they also found the original ASRS v1.1 is able to effectively detect ADHD using the DSM‐5 criteria. In the current study, we use the ASRS v1.1 and the original DSM‐IV scoring rule; it is unclear if results will be similar when researchers begin to use the new measure.

Despite these limitations and the need for more research on item weighting, the current study suggests that the ASRS Screener is promising as a reliable and valid screener for ADHD in adolescents. In particular, the finding that reliability and validity results are as strong (or, in some cases, stronger) among the youngest adolescents in the sample, suggests that the measure might be useful among youth as young as 11 or 12 years old. Given the brevity of this screener, inclusion in surveillance systems might be feasible, where it has traditionally not been for longer measures of ADHD symptoms, and therefore provide opportunities to more effectively identify and respond to ADHD among adolescents.

Green JG, DeYoung G, Wogan ME, Wolf EJ, Lane KL, Adler LA. Evidence for the reliability and preliminary validity of the Adult ADHD Self‐Report Scale v1.1 (ASRS v1.1) Screener in an adolescent community sample. Int J Methods Psychiatr Res. 2019;28:e1751 10.1002/mpr.1751

Dr. Adler received grant and research support in the past year from Sunovion Pharmaceuticals, Enzymotec, Shire Pharmaceuticals, and Lundbeck; served as a consultant for Sunovion Pharmaceuticals, Shire Pharmaceuticals, Otsuka Pharmaceuticals, Bracket, National Football League, and Major League Baseball; and has received royalty payments (as inventor) from New York University for license of adult attention‐deficit/hyperactivity disorder (ADHD) scales and training materials since 2004.

REFERENCES

  1. Adler, L.A. , Kessler, R.C. , Spencer, T. (2003). Adult ADHD self‐report scale‐V1.1 (ASRS‐V1.1) screener from WHO composite international diagnostic interview. Available at: https://www.hcp.med.harvard.edu/ncs/ftpdir/adhd/6Q_ASRS_English.pdf. Accessed November 28, 2017.
  2. Adler, L. A. , Shaw, D. M. , Spencer, T. J. , Newcorn, J. H. , Hammerness, P. , Sitt, D. J. , … Faraone, S. V. (2012). Preliminary examination of the reliability and concurrent validity of the attention‐deficit/hyperactivity disorder self‐report scale V1. 1 symptom checklist to rate symptoms of attention‐deficit/hyperactivity disorder in adolescents. Journal of Child and Adolescent Psychopharmacology, 22, 238–244. 10.1089/cap.2011.0062 [DOI] [PubMed] [Google Scholar]
  3. Adler, L. A. , Spencer, T. , Faraone, S. V. , Kessler, R. C. , Howes, M. J. , Biederman, J. , & Secnik, K. (2006). Validity of pilot adult ADHD Self‐Report Scale (ASRS) to rate adult ADHD symptoms. Annals of Clinical Psychiatry, 18, 145–148. 10.1080/10401230600801077 [DOI] [PubMed] [Google Scholar]
  4. American Psychiatric Association (1994). Diagnostic and statistical manual of mental disorders (4th ed.). Washington, DC: American Psychiatric Association. [Google Scholar]
  5. American Psychiatric Association (2013). Diagnostic and statistical manual of mental disorders (5th ed.). Arlington, VA: American Psychiatric Publishing; 10.1176/appi.books.9780890425596 [DOI] [Google Scholar]
  6. Biederman, J. , Mick, E. , & Faraone, S. V. (2000). Age‐dependent decline of symptoms of attention deficit hyperactivity disorder: Impact of remission definition and symptom type. American Journal of Psychiatry, 157, 816–818. 10.1176/appi.ajp.157.5.816 [DOI] [PubMed] [Google Scholar]
  7. Breslau, J. , Miller, E. , Breslau, N. , Bohnert, K. , Lucia, V. , & Schweitzer, J. (2009). The impact of early behavior disturbances on academic achievement in high school. Pediatrics, 123, 1472–1476. 10.1542/peds.2008-1406 [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Breslau, J. , Miller, E. , Chung, W. J. J. , & Schweitzer, J. B. (2011). Childhood and adolescent onset psychiatric disorders, substance use, and failure to graduate high school on time. Journal of Psychiatric Research, 45, 295–301. 10.1016/j.jpsychires.2010.06.014 [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Brown, T. (2006). Confirmatory factor analysis for applied research. New York, NY: Guilford. [Google Scholar]
  10. Caci, H. M. , Morin, A. J. S. , & Tran, A. (2014). Prevalence and correlates of attention deficit hyperactivity disorder in adults from a French community sample. Journal of Nervous and Mental Disease, 202, 324–332. 10.1097/NMD.0000000000000126 [DOI] [PubMed] [Google Scholar]
  11. Conners, C. K. , Sitarenios, G. , Parker, J. D. A. , & Epstein, J. N. (1998). Revision and restandardization of the Conners Teacher Rating Scale (CTRS‐R): Factor structure, reliability, and criterion validity. Journal of Abnormal Child Psychology, 26, 279–291. 10.1023/A:1022606501530 [DOI] [PubMed] [Google Scholar]
  12. Cuffe, S. P. , Moore, C. G. , & McKeown, R. (2009). ADHD and health services utilization in the national health interview survey. Journal of Attention Disorders, 12(4), 330–340. 10.1177/1087054708323248 [DOI] [PubMed] [Google Scholar]
  13. Dowdy, E. , Ritchey, K. , & Kamphaus, R. W. (2010). School‐based screening: A population‐based approach to inform and monitor children's mental health needs. School Mental Health, 2, 166–176. 10.1007/s12310-010-9036-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Duckworth, A. L. , & Kern, M. L. (2011). A meta‐analysis of the convergent validity of self‐control measures. Journal of Research in Personality, 45(3), 259–268. 10.1016/j.jrp.2011.02.004 [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Faries, D. E. , Yalcin, I. , Harder, D. , & Heiligenstein, J. H. (2001). Validation of the ADHD rating scale as a clinician administered and scored instrument. Journal of Attention Disorders, 5, 107–115. 10.1177/108705470100500204 [DOI] [Google Scholar]
  16. Fuller‐Killgore, M. D. , Burlison, J. , & Dwyer, W. (2012). Comparison of three ADHD screening instruments in college students of varying cognitive ability. Journal of Attention Disorders, 17, 449–454. [DOI] [PubMed] [Google Scholar]
  17. Goodman, R. (2001). Psychometric properties of the strengths and difficulties questionnaire. Journal of the American Academy of Child & Adolescent Psychiatry, 40, 1337–1345. 10.1097/00004583-200111000-00015 [DOI] [PubMed] [Google Scholar]
  18. Goodman, R. , Ford, T. , Simmons, H. , Gatward, R. , & Meltzer, H. (2000). Using the Strengths and Difficulties Questionnaire (SDQ) to screen for child psychiatric disorders in a community sample. The British Journal of Psychiatry, 177, 534–539. 10.1192/bjp.177.6.534 [DOI] [PubMed] [Google Scholar]
  19. Goodman, R. , & Scott, S. (1999). Comparing the strengths and difficulties questionnaire and the child behavior checklist: Is small beautiful? Journal of Abnormal Child Psychology, 27(1), 17–24. 10.1023/A:1022658222914 [DOI] [PubMed] [Google Scholar]
  20. Hart, E. L. , Lahey, B. B. , Loeber, R. , Applegate, B. , & Frick, P. J. (1995). Developmental change in attention‐deficit hyperactivity disorder in boys: A four‐year longitudinal study. Journal of Abnormal Child Psychology, 23, 729–749. 10.1007/BF01447474 [DOI] [PubMed] [Google Scholar]
  21. Howell, D. C. (2010). Statistical methods for psychology (7th ed.). Belmont, CA: Cengage Wadsworth. [Google Scholar]
  22. Hu, L. , & Bentler, P. M. (1999). Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Structural Equation Modeling: A Multidisciplinary Journal, 6(1), 1–55. [Google Scholar]
  23. Kaufman, J. , Birmaher, B. , Brent, D. , Rao, U. , Flynn, C. , Moreci, P. , … Ryan, N. (1997). Schedule for affective disorders and schizophrenia for school‐age children‐present and lifetime version (K‐SADS‐PL): Initial reliability and validity data. Journal of the American Academy of Child & Adolescent Psychiatry, 36, 980–988. 10.1097/00004583-199707000-00021 [DOI] [PubMed] [Google Scholar]
  24. Keeley, L. M. , Makol, B. A. , Qasmieh, N. , Deros, D. E. , Karp, J. N. , Lipton, M. F. , … De Los Reyes, A. (2018). Validity of adolescent and parent reports on the six‐item ADHD Self‐Report Scale (ASRS‐6) in clinical assessments of adolescent social anxiety. Journal of Child and Family Studies, 27, 1041–1053. 10.1007/s10826-017-0950-y [DOI] [Google Scholar]
  25. Kent, K. M. , Pelham, W. E. , Molina, B. S. , Sibley, M. H. , Waschbusch, D. A. , Yu, J. , … Karch, K. M. (2011). The academic experience of male high school students with ADHD. Journal of Abnormal Child Psychology, 39(3), 451–462. 10.1007/s10802-010-9472-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Kessler, R. C. , Adler, L. A. , Ames, M. , Demler, O. , Faraone, S. , Hiripi, E. , … Walters, E. E. (2005). The World Health Organization Adult ADHD Self‐Report Scale (ASRS): A short screening scale for use in the general population. Psychological Medicine, 35, 245–256. 10.1017/S0033291704002892 [DOI] [PubMed] [Google Scholar]
  27. Kessler, R. C. , Adler, L. A. , Gruber, M. J. , Sarawate, C. A. , Spencer, T. , & Van Brunt, D. L. (2007). Validity of the World Health Organization Adult ADHD Self‐Report Scale (ASRS) Screener in a representative sample of health plan members. International Journal of Methods in Psychiatric Research, 16, 52–65. 10.1002/mpr.208 [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Kessler, R. C. , Avenevoli, S. , Costello, E. J. , Georgiades, K. , Green, J. G. , Gruber, M. J. , … Merikangas, K. R. (2012). Prevalence, persistence, and sociodemographic correlates of DSM‐IV disorders in the National Comorbidity Survey Replication Adolescent Supplement. Archives of General Psychiatry, 69, 372–380. 10.1001/archgenpsychiatry.2011.160 [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Kessler, R. C. , Green, J. G. , Adler, L. A. , Barkley, R. A. , Chatterji, S. , Faraone, S. V. , … Van Brunt, D. L. (2010). Structure and diagnosis of adult attention‐deficit/hyperactivity disorder: Analysis of expanded symptom criteria from the Adult ADHD Clinical Diagnostic Scale. Archives of General Psychiatry, 67, 1168–1178. 10.1001/archgenpsychiatry.2010.146 [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. McCandless, S. , & Laughlin, L. O. (2007). The clinical utility of the Behavior Rating Inventory of Executive Function (BRIEF) in the diagnosis of ADHD. Journal of Attention Disorders, 10(4), 381–389. 10.1177/1087054706292115 [DOI] [PubMed] [Google Scholar]
  31. McLeod, J. D. , Uemura, R. , & Rohrman, S. (2012). Adolescent mental health, behavior problems, and academic achievement. Journal of Health and Social Behavior, 53, 482–497. 10.1177/0022146512462888 [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. McNeish, D. (2017). Thanks coefficient alpha, we'll take it from here. Psychological Methods. 10.1037/met0000144 [DOI] [PubMed] [Google Scholar]
  33. Merikangas, K. R. , He, J. P. , Burstein, M. , Swanson, S. A. , Avenevoli, S. , Cui, L. , … Swendsen, J. (2010). Lifetime prevalence of mental disorders in US adolescents: Results from the National Comorbidity Survey Replication–Adolescent Supplement (NCS‐A). Journal of the American Academy of Child & Adolescent Psychiatry, 49, 980–989. 10.1016/j.jaac.2010.05.017 [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Morin, A. J. , Tran, A. , & Caci, H. (2016). Factorial validity of the ADHD Adult Symptom Rating Scale in a French community sample. Journal of Attention Disorders, 20(6), 530–541. 10.1177/1087054713488825 [DOI] [PubMed] [Google Scholar]
  35. Muthén, L. K. , & Muthén, B. O. (1998). ‐2017). Mplus user's guide (Eighth ed.). Los Angeles, CA: Muthén & Muthén. [Google Scholar]
  36. Sijtsma, K. (2009). On the use, the misuse, and the very limited usefulness of Cronbach's alpha. Psychometrika, 74, 107–120. 10.1007/s11336-008-9101-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Silverstein, M. J. , Alperin, S. , Faraone, S. V. , Kessler, R. C. , & Adler, L. A. (2017). Online First). Test‐retest reliability of the Adult ADHD Screening Scale (ASRS) v1.1 Screener in non‐ADHD controls from a primary care physician practice. Family Practice, 35(3), 336–341. [DOI] [PubMed] [Google Scholar]
  38. Sivertsen, B. , Harvey, A. G. , Pallesen, S. , & Hysing, M. (2015). Mental health problems in adolescents with delayed sleep phase: Results from a large population‐based study in Norway. Journal of Sleep Research, 24, 11–18. 10.1111/jsr.12254 [DOI] [PubMed] [Google Scholar]
  39. Sivertsen, B. , Skogen, J. C. , Jakobsen, R. , & Hysing, M. (2015). Sleep and use of alcohol and drug in adolescence. A large population‐based study of Norwegian adolescents aged 16 to 19 years. Drug and Alcohol Dependence, 149, 180–186. 10.1016/j.drugalcdep.2015.01.045 [DOI] [PubMed] [Google Scholar]
  40. Skogen, J. C. , Sivertsen, B. , Lundervold, A. J. , Stormark, K. M. , Jakobsen, R. , & Hysing, M. (2014). Alcohol and drug use among adolescents: And the co‐occurrence of mental health problems. Ung@hordaland, a population‐based study. BMJ Open, 4, 1–10. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Sonnby, K. , Skordas, K. , Olofsdotter, S. , Vadlin, S. , Nilsson, K. W. , & Ramklint, M. (2015). Validation of the World Health Organization Adult ADHD self‐report scale for adolescents. Nordic Journal of Psychiatry, 69(3), 216–223. 10.3109/08039488.2014.968203 [DOI] [PubMed] [Google Scholar]
  42. Steiger, J. H. (1980). Tests for comparing elements of a correlation matrix. Psychological Bulletin, 87, 245–251. 10.1037/0033-2909.87.2.245 [DOI] [Google Scholar]
  43. Tavakol, M. , & Dennick, R. (2011). Making sense of Cronbach's alpha. International Journal of Medical Education, 2, 53–55. 10.5116/ijme.4dfb.8dfd [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Thomas, S. A. , Daruwala, S. E. , Goepel, K. A. , & De Los Reyes, A. (2012). Using the Subtle Avoidance Frequency Examination in adolescent social anxiety assessments. Child & Youth Care Forum, 41, 547–559. 10.1007/s10566-012-9181-y [DOI] [Google Scholar]
  45. Ustun, B. , Adler, L. A. , Rudin, C. , Faraone, S. V. , Spencer, T. J. , Berglund, P. , … Kessler, R. C. (2017). The World Health Organization adult attention‐deficit/hyperactivity disorder self‐report screening scale for DSM‐5. JAMA Psychiatry, 74, 520–526. 10.1001/jamapsychiatry.2017.0298 [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Wolraich, M. L. , Lambert, W. , Doffing, M. A. , Bickman, L. , Simmons, T. , & Worley, K. (2003). Psychometric properties of the Vanderbilt ADHD diagnostic parent rating scale in a referred population. Journal of Pediatric Psychology, 28, 559–568. 10.1093/jpepsy/jsg046 [DOI] [PubMed] [Google Scholar]
  47. Wolraich, M. L. , Wibbelsman, C. J. , Brown, T. E. , Evans, S. W. , Gotlieb, E. M. , Knight, J. R. , … Wilens, T. (2005). Attention‐deficit/hyperactivity disorder among adolescents: A review of the diagnosis, treatment, and clinical implications. Pediatrics, 115, 1734–1746. 10.1542/peds.2004-1959 [DOI] [PubMed] [Google Scholar]

Articles from International Journal of Methods in Psychiatric Research are provided here courtesy of Wiley

RESOURCES