****** LDQ*****.****/****************Ze

ntall and BeikeLearning Disability Quarterly

Learning Disability Quarterly

Achievement of Social Goals

35(1) 39 53

Hammill Institute on Disabilities 2012

Reprints and permission: http://www.

of Younger and Older Elementary sagepub.com/journalsPermissions.nav

DOI: 10.1177/0731948711429009

http://ldq.sagepub.com

Students: Response to Academic

and Social Failure

Sydney S. Zentall1 and Suzanne M. Beike1

Abstract

Children with mild disabilities experience sufficient failure to produce negative future expectations (goals), which may

compound early academic and social deficits. This research compared the teacher- and student-rated goals of 57 children

at two age levels, who were average learners, had a reading problem/disability (RP), and were hyperactive or had ADHD

(H). In line with predictions of academic impairment, the findings were that students with RP were teacher-rated more

than the H group as less motivated to do their best (work avoidant), especially at older age levels, with fewer goals

related to self-determination. Students with H also reported fewer preferences to work well with others and preferred

to work independently (social avoidance goals), in line with their social impairment (more than the RP group). Educational

implications are presented for identification, for timing of intervention, and for differentiating instruction.

Keywords

ADHD, assessment, at risk, reading, social skills

The ability to read in the first grade is directly related to interventions must be employed in general education prior

long-term outcomes in student achievement (reading com- to a disability diagnosis (i.e., the eligibility determination

prehension, vocabulary, and general knowledge), as assessed procedure known as Response to Intervention, or RtI). The

in a 10-year follow-up study even when partialing out cog- assumption underlying the law is that inadequate instruction

nitive ability (Cunningham & Stanovich, 1997). This is cannot be a reason for diagnosis (20 U.S.C. 1414).

understood by educators, who see that reading can contrib- Although each definition will draw different samples of

ute more generally to performance in academic areas that children, what is common across these different definitions

require reading, such as math, science, and social studies. is early failure in reading. In the current definition, respon-

The importance of reading is magnified by evidence that a sibility for failure is now placed more centrally within the

reading disorder is one of the most common diagnoses in instructional context, even though reading disorder is

school-age children (Bental & Tirosh, 2007). Students widely accepted to be a within-child skill deficit in phone-

achieving below age or grade level, but who are undiag- mic awareness (a skill that underlies decoding abilities).

nosed, add even more students to the current prevalence Another group of students represents the second most

rates of school-age children with reading disabilities (RD; common diagnosis in school-age children attention-

achieving below IQ expectancy; 5 17% prevalence rates; deficit/hyperactivity disorder (ADHD; Bental & Tirosh,

Shaywitz & Shaywitz, 2001). Recent evidence suggests that 2007). These students with ADHD have also been variously

these diagnosed and undiagnosed students are comparable labeled hyperactive, attention problem, attention-deficit

in skill deficits and in their response to intervention (e.g.,

Shaywitz, Morris, & Shaywitz, 2008). In addition to the

1

identification of these two groups, the current diagnosis of Purdue University, West Lafayette, IN, USA

reading disability has changed in response to the 2004 reau-

Corresponding Author:

thorization of the Individuals with Disabilities Education Sydney S. Zentall, Department of Educational Studies, Purdue University,

Act (IDEA), retitled as the Individuals with Disabilities Edu- Beering Hall, West Lafayette, IN 47907, USA

cation Improvement Act. This law requires that levels of Email: abp8gu@r.postjobfree.com

40 Learning Disability Quarterly 35(1)

disorder (ADD), hyperactive/impulsive subtype, inattentive intrinsic to an extrinsic orientation (Vansteenkiste, Lens, &

subtype, and at risk for ADHD. Regardless of the different Deci, 2006).

labeling and levels of severity, all children within this gen- Research examining response to failure has typically

eral category are interrelated theoretically in their perfor- assessed immediate response to academic failure. For exam-

mance deficits documented under specific low stimulus ple, students with ADHD respond to failure immediately with

conditions (i.e., long, repetitive, detailed/complex, or delay a helpless style (e.g., quitting earlier than their peers), with

tasks; for theory and review, see Zentall, 2005b; Zentall & greater intensity of emotional responses and greater than nor-

Zentall, 1983). Documented even more frequently than aca- mal sensitivity to positive and negative social feedback (for

demic failure is the social failure of children with ADHD, reviews, see Dunn & Shapiro, 1999; Zentall, 2005a). Those

who are disliked by peers in general education classes (e.g., few studies that examine response to a longer history of fail-

Madan-Swain & Zentall, 1990), and those children who ure document that students with ADHD more than typical

have been clinically labeled are rejected by peers (e.g., students report fewer social participatory goals and more per-

Mrug, Hoza, Pelham, Gnagy, & Greiner, 2007). Even a formance avoidance goals, and they endorse performance

brief encounter with an unfamiliar student with ADHD is goals more than learning goals (7 to 12-year-olds, Dunn &

sufficient to occasion rejection by typical peers (e.g., Shapiro, 1999; 10 to 13-year-olds, Barron et al., 2006). The

Pelham & Bender, 1982). This failure to establish and main- only other assessments across a span of time have involved a

tain positive peer relationships has been linked to long-term comparison of ADHD subtypes (Hyperactive/Impulsive vs.

negative outcomes across multiple studies (for review, see Inattentive ADHD, or high aggressive vs. low aggressive

King et al., 2009). ADHD; Carlson, Booth, Shin, & Canu, 2002; Melnick &

Taken together, we propose that failure is common to Hinshaw, 1996).

both of these mild disability areas and is likely to have con- In the field of learning disabilities (LD), students with LD

sequences in the types of goals that students subsequently also gave up more easily and had a helpless learning style,

endorse. Goals are essentially what gives activity purpose more negative emotions (e.g., confusion, anxiety, frustration),

or meaning (Maehr & Zusho, 2009, p. 77). Goal selection and higher levels of work avoidance prior to and immediately

has been examined within a number of theoretical frame- following failure on an academic task compared to students

works. Traditional theory has examined achievement moti- without disabilities (e.g., Nelson & Manset-Williamson,

vation within the confines of two major goals mastery and 2006; Sideridis, 2005). In response to a longer history of

performance. Performance goals are held by individuals failure, disability comparisons have been made with typically

who are interested in receiving positive evaluations about developing comparisons at one age level. This literature has

their performance in relation to normative comparisons. For consistently documented that students with LD have an exter-

example, individuals with performance goals may be moti- nal locus of control but without finding a clear pattern of goal

vated to present themselves in a positive light and as having orientations that differs from typical comparisons (for review,

superior skills and ability evaluated in comparison to others see Sideridis, 2009).

(extrinsic motivation; Nicholls, 1984; Ohan & Johnston, Thus, prior research has primarily examined immediate

2002). It is unfortunate that performance goals can be prob- response to academic failure or contrasted students with and

lematic when children are placed in difficult academic situ- without disability. These disability versus nondisability

ations. Under these conditions, children focus on avoiding comparisons fail to provide us with sufficient information

incompetence, rather than gaining competence. On the about what characterizes one group as different from another

other hand, when students endorse mastery goals, it indi- mild disability group. As stated by Wong (1986) about the

cates intrinsic motivation to engage in an activity (Schunk, field of LD, cumulative research of LD and non-LD chil-

Pintrich, & Meece, 2008). Individuals with these goals seek dren becomes almost redundant since the former children

academic challenges and demonstrate persistence, which invariably give inferior performances, compared to non-LD

are significant predictors of adaptive academic outcomes children (pp. 18 19).

(Barron, Evans, Baranik, Serpell, & Buvinger, 2006). Thus, advances in the field at this time may be better

Goal theory has been further divided into classes of served by an assessment of whether a history of failure pre-

goals achievement and social (Maehr & Zusho, 2009) dicts goal outcomes that are specific to type of failure (dis-

which have been examined in relation to basic psychological ability group) and whether these goal outcomes change

needs for autonomy, competence, and relatedness (i.e., self- over time (e.g., generalize or accumulate). To this purpose

determination theory; Ryan & Deci, 2009). Self-determination in this study, we were interested in recruiting students from

theory proposes that human beings are inherently proactive two different disability groups and assessing how they dif-

and begin life with a generally high level of intrinsic motiva- fered from one another and from typically developing class-

tion (Deci & Ryan, 1985). However, when academic or social mates. This provided an advance over past research because

learning failures occur, there is more external (adult) pressure (a) all participants were selected from the same schools

to do well, which can alter a child s motivation from an (i.e., rather than recruiting clinical and nonclinical samples

41

Zentall and Beike

using different procedures and sources), (b) ratings were Woodcock-Johnson Test of Achievement III (Woodcock,

made by the students and by their teachers, and (c) two age McGrew, & Mather, 2001).

levels were assessed. We assumed that the effects of failure We combined students with school-defined RD and

would be specific at early age levels and generalize across reading difficulties into one group called RP, at risk for RD,

areas over time (i.e., only appear at the older age levels). We because there were no differences between the diagnosed

predicted that students with reading failure would endorse and non-diagnosed groups in STAR reading, PPVT-III

fewer self-determination achievement goals (greater reli- scores, or Conners ratings (see below) and because research

ance on others and on external standards, with an avoidance has consistently demonstrated that readers achieving below-

of challenge and of work) relative to the other groups. In IQ expectancy and readers achieving below age or grade

contrast, students with behavioral problems and associated level are comparable and different from normal-achieving

social failure would endorse fewer social goals (less reli- readers (Shaywitz et al., 2008).

ance on others, less cooperation, more social isolation or At risk for impairment is a necessary condition for dis-

social avoidance) especially over time than average learners ability diagnosis, and most of these students were receiving

or students with academic failure. extra support (e.g., multi-tiered instructional support) in a

Response to Intervention diagnostic system. That is, all of

these students scored at least 6 months below grade level

Method expectancy on the STAR Reading Test and had been placed

Participants in one of three tiered intervention categories. For example,

at the third grade, Tier 1 accommodations were provided

Fifty-seven children (6 to 11 years in age) were recruited through differentiated instruction for students who fell

from a suburban Midwestern school of approximately 500 below the 40th percentile or below a scaled score of 310.

students, with a mean household income of approximately Tier 2 interventions provided was below the 30th percen-

$50,000. The percentage of students eligible for free lunch tile and a scaled score of 257. Tier 3 urgent interventions

was 31%, similar to the state s average of 38%. To assess needed was below the 10th percentile and a scaled score of

single-word receptive vocabulary, each child was individu- 180. For example, a student in third grade who scored at a

ally administered Form A of the Peabody Picture Vocabulary grade equivalent of 2.5 would be at the 32nd percentile,

Test (PPVT-III Form A [test retest = .91 .94]; Dunn & with a standard score of 282, and typically would receive

Dunn, 1997) and then recruited using a criterion cutoff score in-class accommodations within Tier 1.

of 85 or higher (i.e., average single-word receptive language Each child for whom we received parental permission had

ability). In each child s file were data from the most recent been initially rated by his or her teacher using the Conners

winter scores on a computerized reading test (STAR Reading Teacher Rating Scale (CTRS-28; Conners, 1989). The CTRS-

Test, version 2.2; Renaissance Learning Place, 2007). For 28 consists of 28 items constructed to reflect the characteris-

this reading test, the reliability coefficients across an interval tics of ADHD as stated in the fourth edition, text revision, of

of 1 week ranged from .79 to .91, and alternate forms reli- the Diagnostic and Statistical Manual of Mental Disorders

ability ranged from .79 to .91 across grade levels. Construct (DSM-IV-TR; American Psychiatric Association, 2000). The

validity between the STAR Reading Test and other standard- CTRS-28 demonstrates good internal and test retest reliabili-

ized tests has yielded correlations between .60 and .90, ties (.88 .95 and .72 .92, respectively; Conners, 1989).

generally around .80. ADHD rating scales have been shown to have specificity

Group status. For our disability groups, we selected stu- greater than 94% in studies differentiating children with

dents in inclusion contexts with mild to severe problems in ADHD from typical age-matched community controls

reading (i.e., reading difficulties and disabilities) and in (American Academy of Pediatrics, 2001). Recent evidence

social/behavioral characteristics (hyperactivity and ADHD). has indicated that these teacher ratings, which are not specific

This procedure of selecting students from a large inclusion to any one day or situation, are moderately to strongly related

school had the advantage of reducing probable co-diagnoses to student behavior recorded by an independent observer over

and setting-specific differences in motivation to participate. 3 to 4 days; this further supports the validity of the Conners

Eighteen children 2 of whom were Hispanic met the scale (Lauth, Heubeck, & Mackowiak, 2006).

criteria of reading problems (RP); 12 of these children were All 17 of these children (2 were Hispanic) had scores in

currently receiving services in special education for LD in the clinical range of 70 or higher (two standard deviations

reading (reading disabled) and the other 6 were at risk for above the mean of 50; SD = 10) on the hyperactive subscale

RD. Children labeled by the school as reading disabled had (i.e., behavioral deficits); 6 of these also scored two stan-

a minimum 18-point score difference between the fifth dard deviations or higher on the Inattentive Passive sub-

edition of the Stanford-Binet Intelligence Test (Roid, 2003) scale, and 3 of these 17 students had a pediatric diagnosis of

and reading subtests (Letter Word Identification, Word Attack, ADHD. We labeled this school-based sample hyperactive

Passage Comprehension, Reading Vocabulary) on the (H) and at risk for ADHD, rather than ADHD, because we

42 Learning Disability Quarterly 35(1)

did not assess age of onset of ADHD symptoms, obtain on lower level skill development (e.g., decoding, calcula-

parental ratings in addition to teacher ratings, assess degree tions) versus higher elementary grades (3 5) that focus on

of impairment, or conduct diagnostic interviews. All of these higher level skills (e.g., reading comprehension, problem

students performed at or above grade level on the STAR solving).

computerized reading test.

All 22 comparison (C) children, of whom 3 were Hispanic,

Motivational Measures and Procedures

scored in the range of 29 to 50 on all subscales of the

Conners scale. Scores on the STAR Reading Test were at or After the researcher assigned students to groups, each

above grade level. teacher, na ve to student group assignment, filled out the

Screening for children in the above three groups pro- scale entitled Teacher Rating of Academic Achievement

duced a sample with parental permission that overrepre- Motivation (TRAAM; Stinnett & Oehler-Stinnett, 1992).

sented males. That is, of the students in the H group, 16 The TRAAM has good internal consistency ( = .79 .98)

were male (1 female); in the RP group, 14 were male (4 and good convergent, discriminative, and predictive validi-

female); and in the C group, 18 were male (4 female). This ties (Stinnett & Oehler-Stinnett, 1992). This questionnaire

could be explained by the greater number of boys versus has 45 items, grouped into six factors, with most items

girls in that school system (58% males; 42% females), by assessing achievement motivation (the first factor). The fac-

the parental permissions received, and by the overidentifi- tors are as follows:

cation of boys with disabilities in reading (2 15 to 1) and

with ADHD (3 6 to 1; Lyon, 1996; Polanczyk & Jensen, T1: Amotivation does the minimum required;

2008; Vogel, 1990; Willcutt, Pennington, Chhabildas, Olson, T2: Mastery is concerned with learning new things

& Hulslander, 2005). Such statistics are approximations for well;

this sample because they typically represent clinically T3: Academic-Cognitive Skills lacks basic cogni-

referred students with co-occurring disabilities. In this study, tive skills (assesses teachers perceptions of the

we recruited samples without co-occurring disabilities to child s cognitive ability);

assess our predictions of different goal outcomes of each T4: Academic Work, Completion completes assign-

group. Furthermore, when students have co-occurring ADHD ment without teacher intervention;

and RD, the findings suggest a magnification of problems T5: Competition looks to be the best on a given aca-

associated with each (Mayes, Calhoun, & Crowell, 2000). demic task; and

That is, children with RD plus ADHD have more problems T6: Cooperation works well with other students

in attention and more difficulty with reading than either (assesses social motivation).

group singly (Lyon, 1996). Thus, a co-occurring group would not

be considered representative of either group, even though Each student participated in two sessions, each approxi-

there are some performance equivalencies between the mately 15 min long. For each session, students were indi-

pure and co-occurring groups (Willcutt et al., 2005). vidually escorted to an empty classroom (7 m by 7 m) with

Table 1 presents demographic data that can be used to 8 tables, each approximately 2 meters long, positioned

compare the three groups to assess whether our defining cri- around the outside perimeter, with three computers against

teria produced significant differences among our groups one wall. The participant sat at the end of one table and the

(i.e., validity). Students in the RP group were significantly researcher sat at the adjacent end, where the tables formed a

lower on STAR Reading Test performance than the compari- corner. The first session involved the individual administra-

sons and than the students in the H group. Furthermore, the tions of the PPVT-III.

C and H groups did not differ in reading performance. The In the second session, Harter s (1980, 1981) scale of

H group was higher on the Conners subscales (hyperactiv- intrinsic versus extrinsic motivational orientation was

ity, inattention, and conduct problems) than the other two orally administered individually to each student. This

groups, who did not differ. The RP group was also lower in questionnaire was selected as an age-appropriate measure

individually assessed single-word language than the H or of general academic motivation (i.e., not a specific aca-

C groups, who did not differ. This difference in single-word demic area, such as reading or math). Other scales were

verbal vocabulary would be expected for students with considered but were rejected because there was no assess-

reading difficulties/disabilities, who have deficits storing ment by teachers (e.g., Elliott & Murayama, 2008; Vallerand

and retrieving abstract verbal information (Ghelani, Sidhu, et al., 1992), it was used for older students (Guay, Vallerand,

Jain, & Tannock, 2004). & Blanchard, 2000), or it was specific to an academic area,

Table 1 also reports the number of students in each age such as reading or math (Gottfried, 1990; Karsenti &

group (6 8 and 9 11). This age breakdown was selected to Thiebert, 1996). The Harter scale has 30 items with five

represent early elementary grades (K Grade 2) that focus subscale variables:

43

Zentall and Beike

Table 1. Demographic Equivalence: Hyperactive (H), Reading Problem (RP), and Comparison (C) Groups

RP5 (n = 18) R2

Demographic H (n = 17) C (n = 22) df MSE F p

Age

In months 103-***-***-** .079 288.20 0.11 .1070

6 8 years (n) 10 7 13

9 11 years (n) 7 11 9

STAR reading

2.68a 1.48b 2.99a

LSM 56 .76 0.35 31.59

t novelty on students with attention and learning problems.

Copyright of Learning Disability Quarterly is the property of Sage Pu

Contact this candidate