Abstract
It has been well established that ADHD leads to impairment in the academic, social, and occupational functioning of college students and young adults (Barkley & Murphy, 2006). College students with ADHD tend to have lower grade point averages (GPA), are more likely to be on academic probation, have more difficulty on tests of reading and writing, and report more academic problems than those without ADHD (Biederman et al., 1993; Heiligenstein, Guenther, Levy, Savino, & Fulwiler, 1999; Roy-Byrne et al., 1997). College students with ADHD also report lower scores in time management, concentration, motivation, anxiety, test-taking skills, and study strategies than those without ADHD (Reaser, Prevatt, Petscher, & Proctor, 2007).
Children with ADHD tend to experience more interpersonal difficulties than those without ADHD, and poor social skills are more common in children with ADHD (Gaub & Carlson, 1997; Hoza, 2007; Kats-Gold, Besser, & Priel, 2007; Thorell & Rydell, 2008). This often leads to a lack of reciprocal friendships and peer unpopularity (Gaub & Carlson, 1997; Hoza, 2007; Nijmeijer et al., 2008; Nixon, 2001). The social difficulties of adults with ADHD are less understood, but some research has shown that college students with ADHD symptoms report lower levels of social skills, more difficulty in recognizing emotion in others, higher rates of expressing anger in socially unacceptable ways, increased emotional dysregulation, and fewer friends compared with their counterparts without ADHD (Rapport, Friedman, Tzelepis, & Van Voorhis, 2002; Reimherr et al., 2007; Shaw-Zirt, Popali-Lehane, Chaplin, & Bergman, 2005).
Occupationally, adults with ADHD are likely to experience a variety of difficulties in the workplace (Painter, Prevatt, & Welles, 2008; Wender, 1995). Wilens, Faraone, and Biederman (2004) have observed that adults with ADHD often have more frequent job changes, lower rates of professional employment, and more work difficulties than do their counterparts without ADHD. Furthermore, Biederman and Faraone (2006) noted that these individuals were significantly less likely to hold full-time employment as compared with a group without ADHD. Employers often find adults with ADHD less adequate in fulfilling work demands, less likely to work independently and complete tasks, and less likely to get along well with supervisors (Barkley & Murphy, 1998).
Despite documented difficulties in the interpersonal, academic, and employment arenas, there is evidence that individuals with ADHD often make very positive judgments about their own performance (e.g., Hoza et al., 2004; Hoza, Pelham, Dobbs, Owens, & Pillow, 2002; Owens, Goldfine, Evangelista, Hoza, & Kaiser, 2007; Owens & Hoza, 2003). This is referred to as the positive illusory bias and is operationally defined as “a disparity between self-report of competence and actual competence such that the self-reported competence is substantially higher than actual competence” (Owens et al., 2007, p. 336). The social psychology literature has long acknowledged that, in general, individuals tend to hold positive illusions regarding themselves, resulting in the effect of most people thinking they are “better than average” (Alicke & Govorun, 2005; Owens et al., 2007; Taylor & Brown, 1988). Taylor and Brown (1988) suggested that these moderate positive illusions are adaptive in that they may serve to enhance one’s motivation, performance, and task persistence.
In the population with ADHD, the positive illusory bias is believed to differ from positive illusions often seen in individuals without ADHD. Although it is common to find that individuals without ADHD have a discrepancy between their perceived and actual abilities (Alicke & Govorun, 2005; Owens et al., 2007; Taylor & Brown, 1988), the discrepancy between the perceived competence and actual competence is greater when the positive illusory bias is in effect (Hoza et al., 2002; Owens et al., 2007; Owens & Hoza, 2003). In addition, although moderate positive illusions may serve as a motivational purpose in normal individuals (Taylor & Brown, 1988), the positive illusory bias that is frequently identified in individuals with ADHD does not lead to increased task persistence and motivation. As studies have indicated, children with ADHD actually tend to give up more frequently on tasks than do their peers without ADHD (Hoza, Pelham, Waschbusch, Kipp, & Owens, 2001; O’Neill & Douglas, 1991; Owens et al., 2007). Thus, the positive illusory bias identified in those with ADHD has been differentiated from the more typical positive illusions found among the population without ADHD. So far, the majority of the research on the positive illusory bias has focused on children. Four studies using adult participants have found that individuals with ADHD demonstrate the positive illusory bias in assessing their ability to recognize facial expressions (Rapport et al., 2002), awareness of their own social competence (Friedman et al., 2003), evaluation of one’s driving behaviors using a driving simulator (Knouse, Bagwell, Barkley, & Murphy, 2005), and giving confidence ratings after completing a time estimation task (Prevatt, Proctor, Baker, Garrett, & Yelland, 2010).
Positive illusory bias in individuals with ADHD has been attributed to a myriad of factors, including cognitive immaturity, neuropsychological deficits, ignorance of one’s own incompetence, and the need for self-protection (Hoza et al., 2002; Milich, 1994; Ohan & Johnston, 2002; Owens et al., 2007; Owens & Hoza, 2003). However, there is some uncertainty as to whether the positive illusory bias is adaptive or maladaptive. On one hand, some degree of positive illusory bias may actually serve to protect one’s motivation and persistence, even in the event of failure. Alternately, Owens et al. (2007) suggested that the positive illusory bias could also be a barrier to individuals’ recognition and acknowledgement that they are struggling and need assistance, thus leading to a decreased motivation for treatment.
In addition to the empirical findings regarding the positive illusory bias, the authors of the current study have anecdotally observed that individuals with ADHD often do not perceive themselves as performing below expectation. In our experience with hundreds of college students referred to our clinic for ADHD evaluations, it is common for these students to maintain that they are an “above-average” employee, driver, or student. This positive rating of overall performance often occurs even in the presence of a litany of examples of poor performance (provided by the client) that might suggest otherwise. It was this observed phenomenon that led us to further investigate whether the positive illusory bias, in combination with executive functioning deficits, was impacting our clients’ self-evaluations.
It was our hypothesis that college students with ADHD tend to give ratings of global behaviors that are inconsistent with their own ratings of very specific behaviors. We hypothesized that when individuals with ADHD are asked to make a very concrete, specific judgment along a single dimension, they can do this relatively accurately. Alternately, when judgments are more global and require sequencing, long-term memory, or inhibition, the judgment will be impaired. We hypothesized that the deficits in executive functioning cause the impaired judgment, whereas the positive illusory bias causes the impaired judgment to err on the positive side. The two areas of functioning that we chose to examine were driving and employment, given that young adults with ADHD tend to exhibit more difficulties in these specific areas than their peers without ADHD (Barkley & Murphy, 1998; Biederman & Faraone, 2006; Wender, 1995; Wilens et al., 2004). We selected areas of functioning that should (a) differentiate between the populations with and without ADHD; (b) include measures of both simple, specific judgments as well as more complex, global judgments; (c) require executive functioning; and (d) be subjective enough to allow the positive illusory bias to emerge. Thus, the following research questions were posed:
Research Question 1: Do college students with ADHD differ from college students without ADHD in their endorsement of specific behavioral information, as measured by their ratings of work and driving behaviors?
Research Question 2: Do college students with ADHD differ from college students without ADHD in their endorsement of global ratings of behavior, as measured by their ratings of work and driving behaviors?
Research Question 3: Are college students with ADHD more likely than college students without ADHD to give ratings indicating a positive illusory bias, as measured by endorsements of global behaviors that are more positive than their ratings of specific behaviors?Specifically, discrepancy scores were created by comparing each participant’s specific rating to his or her global rating. We then evaluated whether these discrepancy scores were greater for the students with ADHD.
Method
Participants
Participants were 197 college students enrolled at a large public university in the Southeastern United States. The group with ADHD (n = 103) were clients who had been seen at an on-campus psychological assessment center and had received a Diagnostic and Statistical Manual of Mental Disorders (4th ed., text rev.; DSM-IV-TR; American Psychiatric Association, 2000) diagnosis of ADHD. The group without ADHD (n = 94) were recruited from general university classes in education, communications, and career development. The sample was 58% female. Seventy-two percent were White, 11% African American, 2% Asian, 11% Hispanic, and 4% unknown. The participants consisted of freshmen (12%), sophomores (16%), juniors (29%), and seniors (34%). The mean age was 21.59 (SD = 3.67). Mean GPA was 3.12 (SD = 0.61) out of 4.0. By group, the demographics were as follows: with ADHD group 56% female compared with 59% for the group without ADHD; group with ADHD mean age 22.42 (SD = 4.77) compared with 20.68 (SD = 1.37) for the group without ADHD; and group with ADHD mean GPA 3.21 (SD = 0.64) compared with 3.03 (SD = 0.55) for the group without ADHD.
Instruments
Work Performance Rating Scale (WPRS)
The WPRS (Barkley & Murphy, 1998) consists of 28 items. The first 18 items assess one’s difficulty with very specific aspects of work performance—for example, “have difficulty sustaining attention in tasks or other work-related activities.” Respondents rate each of the 18 items on a 0 to 3 scale, with the following anchors: never or rarely, sometimes, often, and very often. These items are summed for a total rating of individual work behaviors ranging from 0 to 54, with higher scores indicating more negative performance. This rating will be referred to as total work items rating. Respondents also give a global rating of their work performance by answering the following single question: “In general, how would you rate your overall work performance and productivity as an employee?” This item is answered on a 1 to 5 scale, with the following anchors: excellent, above average, average, below average, and poor, with higher ratings indicating more negative performance. This rating will be referred to as the global work rating. Ten additional items on the WPRS address areas of functioning (coworkers, supervisors, punctuality, etc.) and were not used in the present study.
Driving Behavior Survey (DBS)
The DBS (Barkley & Murphy, 1998) consists of 26 items that assess one’s competency with very specific driving skills—for example, “I drive at a rate of speed that is within the posted speed limits” or “I use turn signals prior to making a turn or changing lanes.” Respondents rate each of the 26 items on a 1 to 4 scale, with the following anchors: not at all or rarely, sometimes, often, and very often. These items are summed for a total rating of individual driving behaviors ranging from 26 to 104, with higher scores indicating more positive performance. This rating will be referred to as total driving skills rating. Respondents also give a global rating of their driving performance by answering the following single question: “Please circle the number that best describes your overall driving performance.” This item is answered on a 1 to 10 scale, with the following anchors: poor, below average, average/satisfactory, above average, and excellent, with higher ratings indicating more positive performance. This rating will be referred to as the global driving rating. In addition, a short survey accompanies the DBS, asking yes/no and open-ended questions about driving history. For this study, the question “How many driving citations have you received?” was utilized.
Procedure
Participants with ADHD had previously undergone a standard evaluation at a university assessment clinic and had given permission for results of their evaluation to be used in future studies. Data from the participants with ADHD consisted of archival data gathered over the previous 3-year period. As part of their original evaluation, clients had completed the following forms, taken from Barkley and Murphy (1998): Current Symptoms Scale, Childhood Symptoms Scale, WPRS, DBS, Employment History Form, and a Social History Form. Clients were also asked to identify an informant (e.g., a parent and roommate) to complete the Current Symptoms Scale–Other Report Form, Childhood Symptoms Scale–Other Report Form, and the WPRS–Other Report Form (Barkley & Murphy, 1998). In addition, clients were asked to complete the Learning and Study Strategies Inventory, 2nd Edition (LASSI; Weinstein & Palmer, 2002). Clients participated in a clinical interview to gather additional data on ADHD symptoms, psychological disorders, developmental history, and family history (Barkley & Murphy, 1998). All clients were administered select cognitive and achievement subtests from the Woodcock–Johnson Tests of Cognitive Abilities and Achievement–III (WJ-III; Woodcock, McGrew, & Mather, 2001) to rule out a learning disability. Clients were diagnosed with ADHD if the following criteria were met (Murphy & Gordon, 2006): (a) there was evidence that the client experienced ADHD symptoms in early childhood (ages 5-12, operationalized as endorsing at least six of nine symptoms of inattention or hyperactivity/impulsivity); (b) there was evidence that, no later than in middle school, these ADHD symptoms led to substantial and chronic impairments across settings; (c) there was evidence that the client currently experienced ADHD symptoms (operationalized as endorsing at least six of nine symptoms of inattention or hyperactivity/impulsivity); (d) there was evidence that these ADHD symptoms currently caused substantial and chronic impairments across settings; and (e) there were no explanations other than ADHD that better accounted for the client’s current symptoms. The group with ADHD reported the following diagnosis subtypes: predominantly inattentive type = 44 (43%), predominantly hyperactive-impulsive type = 4 (4%), combined type = 51 (50%), and unknown = 4 (4%).
Participants without ADHD were recruited from classes in education, communications, and career development, and received extra credit in return for participation. All participants without ADHD learned of the study during class time and were given a web address to complete the DBS, WPRS, and a demographic information form online. Participants were asked to complete the online instruments within 1 week, to receive extra credit for the course. The demographic information form inquired as to whether the participant had ever received a diagnosis of ADHD, and participant’s data were not utilized if they responded affirmatively. Four potential participants were dropped for this reason.
Results
Analysis of Mean Score Differences on Specific and Global Ratings
A one-way analysis of variance was used to compare the participants with ADHD with the participants without ADHD on three ratings of specific behaviors (total work items rating, total driving skills rating, and driving citations) as well as two ratings of global behaviors (global work rating and global driving rating). There were significant between-group differences on all variables except the total driving skills rating. As can be seen in Table 1, participants with ADHD gave self-ratings of specific work behaviors, number of driving citations, global work behaviors, and global driving ratings that were more negative than the participants without ADHD. All were large effect sizes, with the exception of global driving.
Group Differences on Specific and Global Ratings of Work and Driving
A higher rating denotes more negative functioning.
Creation of Specific-to-Global Rating Variables
Variables measuring the relationship between specific and global ratings were created using the DBS and the WPRS. Three separate specific-to-global rating variables were created (one measuring work and two measuring driving), all using the same process. These variables compared ratings of specific behaviors to ratings of global functioning to create a score that indicated whether participants gave a global rating that was more positive than their rating based on specific behaviors. It was believed that a rating change in the positive direction (rating oneself more positively on the global rating than on the specific items) would reflect the positive illusory bias. As the ratings of specific behaviors and the ratings of global function used different metrics, the first step involved transforming each of these ratings to a type of standardized discrepancy score based on frequencies. This method was partially modeled after a discrepancy score used by Knouse et al. (2005) in their study of driving behaviors. For our calculations, scores on the total work items rating (range = 0-54) were recoded into a 1 to 3 scale based on frequencies of the original scale (e.g., approximately the bottom third of the distribution [scores 1-7] were recoded as a 3, approximately the middle third of the distribution [scores 8-29] were recoded as a 2, and approximately the top third of the distribution [scores 30-54] were recoded as a 1). The same procedure was used for the global work rating (range = 1-5), with thirds of the original distribution recoded into scores of 1 to 3. For all transformed scores, a higher score indicated more positive functioning. As a result of these transformations, each participant had a 1 to 3 score for ratings of specific behaviors and a 1 to 3 score for global ratings. Next, a variable was created based on how the participant’s score changed when comparing the specific behaviors to the global ratings. For example, if participants had the same score on the specific and the global ratings (e.g., 1, 1 or 2, 2 or 3, and 3), then their change score would be a “no change.” If they went from a low score on the specific behaviors to a higher score on the global ratings (e.g., from 1 to 2 or 1 to 3 or 2 to 3) then this would indicate a positive illusory bias. Finally, if they went from a high score on the specific behaviors to a lower score on the global rating (e.g., from 3 to 2 or 3 to 1 or 2 to 1) then this would indicate a negative change. Thus, their final change score for work behaviors had three possible values: 1 = no change, 2 = positive change (which would be an indication of the positive illusory bias), and 3 = negative change.
Using the same process, a change score for driving behaviors was created using the total driving skills rating and the global driving rating. Finally, a second driving change score was created using the number of individual citations the participant had received as the measure of specific behaviors and comparing this to the global driving rating. That is, we had two measures of specific-to-global ratings for driving, one comparing individual driving behaviors to global driving ratings and the second comparing number of driving citations to global driving ratings.
Analysis of Specific-to-Global Rating Variables
Chi-square analyses were used to evaluate differences between participants with and without ADHD on the three specific-to-global rating (change score) variables. For the work change score, the χ2 was significant and in the expected direction, χ2(2, 188) = 10.78, p < .005. Thirty-two percent of the participants with ADHD had a positive change score compared with only 12% of the control group. For the driving behaviors change score, the χ2 was not significant (p > .05). Finally, for the driving citations change score, the χ2 was significant and in the expected direction, χ2(2, 166) = 6.56, p < .03. Forty-six percent of the participants with ADHD had a positive change score compared with 31% of the control group. These results can be seen in Table 2.
Group Differences on Specific-to-Global Ratings Based on Chi-Square Analyses
Note: The column “positive change” indicates a rating on the global measure that is more positive than the rating on the specific measure.
Discussion
Participants with ADHD gave self-ratings indicating poorer performance than participants without ADHD on four of five ratings of work and driving behavior. This might suggest that the positive illusory bias is not indicated in this group of college students with ADHD. However, it was our hypothesis that the positive illusory bias might be masked when looking at group differences in absolute functioning. Rather, we were interested in whether those with ADHD gave ratings of global behavior that were more positive than their own ratings of specific behaviors, with regard to the same area of functioning. We found support for this hypothesis in both work and driving. Using a discrepancy score that calculated the change in one’s relative position, we created a measure that compared the specific and global ratings of behaviors. A positive change score indicated that the individual gave an estimate of global behavior that was higher than would be expected, given one’s estimate of specific behaviors. We predicted that estimates of specific behaviors were easy to make and did not require complex executive functioning. Examples of specific questions would involve how many driving citations had been received, whether turn signals were used, or whether one was distracted at work. Alternately, the two global ratings required the individual to consider all the multiple aspects of work and driving, and make a global judgment of overall functioning. These global ratings involved responses such as excellent, average, or poor. In our clinical experience with students with ADHD, it appeared to us that despite making numerous individual ratings of poor performance, students often gave global ratings that indicated they were a good driver or a good employee. We found that the mean rating of the group with ADHD for global functioning for driving and work was in the “above-average” category. We hypothesized that there was a discrepancy between ratings of specific behaviors and ratings of global functioning. We were able to confirm empirically that the students with ADHD did in fact follow this discrepancy pattern. With regard to work behaviors, 32% of the participants with ADHD (compared with 12% of the group without ADHD) gave a global rating of their performance that indicated an overestimate based on what would be predicted from their specific ratings. With regard to driving citations, 46% of participants with ADHD (compared with 31% of the group without ADHD) gave a global rating of their performance that indicated an overestimate based on what would be predicted from their specific ratings. Although some students without ADHD engaged in this pattern, the numbers who did so were significantly smaller than in the group with ADHD.
Future Research
The current study lays the groundwork for future research by demonstrating a pattern of functioning that is consistent with a positive illusory bias. The next step in this line of research would be to empirically test the pattern of functioning while including specific measures of executive functioning. We hypothesized that our findings were the result of two interrelated mechanisms. First, due to possible executive functioning deficits in working memory, sequencing, and inhibition, such as those described by Barkley (1997) and Weyandt (2009), we thought that the students with ADHD would make errors in the global ratings. We speculated that individuals without such deficits might say to themselves, “I just rated myself as having received eight driving citations; therefore, I should give an overall rating that indicates I am average or below average.” However, a person with executive functioning deficits might err in this logic. He or she might have difficulty holding the first set of information (number of citations) in memory as well as a limited ability to analyze that information. Second, the positive illusory bias would predict that when errors are made, they are made in the positive direction. We think it may be the two mechanisms working in conjunction that is important. Otherwise, if only the positive illusory bias in general were operating, we would have found that the group with ADHD gave inflated ratings compared with the group without ADHD on the individual measures. But this was not the case. It was only when comparing the change scores from specific to global that we found the predicted pattern. Future research can explore this speculation using standardized measures of executive functioning. In addition, future research would benefit from expanding the areas of functioning studied to include organizational skills, study habits, and time management skills.
A second phenomenon that we have noticed anecdotally involves asking college students with ADHD why they are late to classes and meetings. When we ask them how long it takes to get to class, they might give an immediate answer that seems somewhat of an underestimate. But when we ask them to list each specific event in their morning routine (showering, getting dressed, eating breakfast, driving to campus, searching for parking, and walking into the building), the estimates for these specific events seem to add up to well more than the original time estimate. This would be an interesting phenomenon to study empirically and to compare with a sample without ADHD. We suspect that the same inability to go from the specific to global would be observed. This would be an interesting follow-up to the current study.
Implications for Practice
Implications of this research for practice would include the suggestion that college students with ADHD may routinely need help with organizing assignments that are more global in nature. One possible intervention might be to encourage students to break tasks into smaller components of simple structure and work through those individual segments. Recent work with college students found that those with ADHD had difficulty with complex tasks that involved numerous steps (Prevatt et al., 2010). Their findings seem to share a similar underlying deficit in dealing with more global, complex thought processes. This suggests that, in general, college students with ADHD need help with apparently routine tasks if they involve executive functioning. In the clinic with which the authors are affiliated, it is common to work with college students on time management, asking them to prepare a schedule for homework, projects, and studying. The current study suggests that the counselor might want to explicitly help the student make scheduling estimates by breaking down the task into its component pieces. Other work on the positive illusory bias (e.g., Owens et al., 2007) suggests that this could also be a barrier to individuals’ recognition and acknowledgment that they are struggling and need assistance, thus leading to a decreased motivation for treatment. An additional implication for treatment is that college students with ADHD might need very specific, structured sessions when working with an academic advisor, as they might be likely to view a series of examples of poor performance or grades, yet conclude that they are doing fine in school. Finally, counselors might work with students to determine whether they are able to make accurate self-analyses of their behavior, such as whether they are performing well in courses. This could be done by first asking the students to make a self-evaluation and then helping them to consider the evidence for that conclusion.
Limitations
The current study is limited in that the measure of positive illusory bias is exploratory and does not yet have empirical evidence of construct validity. Future research needs to corroborate this method of investigating the positive illusory bias. It was hypothesized that the students with ADHD would have executive functioning deficits in memory and analysis to a greater degree than the students without ADHD. However, this assumption was based on past research and was not verified in the present sample. Our sample of students with ADHD evidenced a gender ratio of 56% female. Although this percentage is high compared with rates for children and adults, it is not unusual for college students (e.g., Blasé et al., 2001; Dupaul et al., 2001; Rabiner, Anastopoulos, Costello, Hoyle, & Swartzwelder, 2007; Sparks, Javorsky, & Philips, 2005). Future research should focus on determining whether gender ratios in college students are actually different among other populations. In addition, our sample of students with ADHD had a higher GPA than might be expected. We hypothesize that this was due to the fact that the students with ADHD were actively seeking help (hence had undergone an evaluation of ADHD) and thus might work harder in general than those students who did not seek help. We also hypothesize that the students with ADHD were more likely to be receiving accommodations that would promote their academic success.
Footnotes
The authors declared no potential conflicts of interests with respect to the authorship and/or publication of this article.
The authors received no financial support for the research and/or authorship of this article.