Printer Friendly
The Free Library
22,725,466 articles and books

Elementary Reading Attitude Survey (ERAS) scores in academically talented students.

For some time, there has been an assumption that the desire to read and reading achievement are related (Ley LEY. This word is old French, a corruption of loi, and signifies law; for example, Termes de la Ley, Terms of the Law. In another, and an old technical sense, ley signifies an oath, or the oath with compurgators; as, il tend sa ley aiu pleyntiffe. Brit. c. 27.  & Trentham, 1987), and several studies have indicated that attitudes toward reading are related to scores on reading achievement tests (Askov & Fishback, 1973; Diamond & Onwuegbuzie, 2001; Swanson, 1982; Walberg & Tsai, 1985). However, McKenna and Kear (1990) observed that recent research in reading assessment was more concerned with comprehension comprehension

Act of or capacity for grasping with the intellect. The term is most often used in connection with tests of reading skills and language abilities, though other abilities (e.g., mathematical reasoning) may also be examined.
 rather than attitude, and argued that reading researchers were ignoring an important aspect in the development of literacy. Attitude toward reading has been defined as students' feelings toward reading, which result in approaching or avoiding reading tasks (Cooter coot·er  
n. Lower Southern U.S.
1. An edible freshwater turtle of the genus Chrysemys.

2. Any of various turtles or tortoises. See Regional Note at goober.
 & Alexander, 1984). In this paper, we examined the reliability and structural validity of scores on the ERAS (McKenna & Kear, 1990), an instrument designed to assess reading attitudes in elementary-aged students. To date, there has been no examination of the psychometric psy·cho·met·rics  
n. (used with a sing. verb)
The branch of psychology that deals with the design, administration, and interpretation of quantitative tests for the measurement of psychological variables such as intelligence, aptitude, and
 properties of this instrument's scores in academically talented students.

The Elementary Reading Attitude Survey

McKenna and Kear (1990, p. 626) reported developing the ERAS, "a public-domain instrument ... [that would] enable teachers to estimate attitude levels efficiently and reliably" in an attempt to increase research on attitudes toward reading. Based on a standardization standardization

In industry, the development and application of standards that make it possible to manufacture a large volume of interchangeable parts. Standardization may focus on engineering standards, such as properties of materials, fits and tolerances, and drafting
 sample of over 18,000 students in grades 1-6 from 95 school districts in 38 states, the 20-item instrument yields three scores: a recreational reading score, an academic reading score, and a total score. McKenna and Kear reported moderate to high internal consistency In statistics and research, internal consistency is a measure based on the correlations between different items on the same test (or the same subscale on a larger test). It measures whether several items that propose to measure the same general construct produce similar scores.  coefficients for ERAS scores as well as evidence of structural validity, and they published normative nor·ma·tive  
Of, relating to, or prescribing a norm or standard: normative grammar.

 standards on the three scores for the six grades they studied.

Allen, Cipielewski, and Stanovich (1992) examined the convergent validity Convergent validity is the degree to which an operation is similar to (converges on) other operations that it theoretically should also be similar to. For instance, to show the convergent validity of a test of mathematics skills, the scores on the test can be correlated with scores  of the ERAS scores in a group of 63 fifth graders. They found that ERAS recreational scores were moderately related to (a) minutes read, (b) book title recognition task scores, (c) children's author recognition task scores, (d) reading scores on an activity preference questionnaire, (e) scores on a reading disposition questionnaire, and (f) two measures of vocabulary. However, scores on the ERAS academic subscale were related only to the reading scores on the activity preference scale. These findings suggested that recreational reading is likely to have a bigger impact on academic performance, perhaps simply because students who enjoy reading for recreation read more material more often.

McKenna, Kear, and Ellsworth (1995) examined the relationship between attitudes toward reading using the ERAS and teacher ratings of students' ability to read in a sample of first through sixth graders. These researchers found that academic, recreational, and total attitude scores correlated cor·re·late  
v. cor·re·lat·ed, cor·re·lat·ing, cor·re·lates
1. To put or bring into causal, complementary, parallel, or reciprocal relation.

 significantly with teachers' ratings of reading ability. Kush Kush: see Cush. , Watkins, McAleer, and Edwards (1995) examined the stability of ERAS scores over a 1-year period in a sample of 289 students in grades 1 to 5. They reported stability coefficients of .43, .36, and .43 for Recreational, Academic, and Total Reading scores, respectively, in spite of in opposition to all efforts of; in defiance or contempt of; notwithstanding.

See also: Spite
 the substantial time interval.

Reading Attitudes in High Achievers

Studies on high-achieving and high-ability students have produced similar results. For example, Kennedy and Halinski (1975) found that secondary students in an accelerated academic track scored significantly higher on a measure of reading attitudes than students in a regular academic track. Similarly, Ley and Trentham (1987) found that 87% of 7th- and 8th-grade students identified as gifted responded with higher than neutral reading attitude ratings on the Mikulecky Behavioral Reading Attitude Measure ([MBRAM MBRAM Modified Bram ]; Mikulecky, 1976). When Ley and Trentham compared mean reading attitude ratings of their gifted sample to participants in a previous study in which the MBRAM had been administered to a heterogeneous Not the same. Contrast with homogeneous.

heterogeneous - Composed of unrelated parts, different in kind.

Often used in the context of distributed systems that may be running different operating systems or network protocols (a heterogeneous network).
 sample of seventh and eighth graders, they found the gifted sample to have significantly more favorable fa·vor·a·ble  
1. Advantageous; helpful: favorable winds.

2. Encouraging; propitious: a favorable diagnosis.

 attitudes toward reading. These findings suggest that students' attitudes toward reading are not only linked to reading achievement but are also related to high achievement in general and to cognitive measures.

Gender and Grade Level Differences in Reading Attitudes

Examinations of reading attitudes have also focused on gender differences. A consistent finding has been that across grades, girls possess more positive attitudes toward reading than boys (e.g., Anderson, Tollefson, & Gilbert, 1985; Diamond & Onwueguzie, 2001; Kennedy & Halinski, 1975; Ley & Trentham, 1987; Ross & Fletcher Fletcher may refer to one of the following: Ideas and companies
  • A fletcher makes arrows, see fletching.
  • Fletcher School of Law and Diplomacy, the graduate school of international relations of Tufts University, located in Medford, Massachusetts.
, 1989; Smith, 1990). Using the ERAS, McKenna et al. (1995) reported that girls in grades 1 to 6 had more positive reading attitudes than boys toward both academic and recreational reading activities. Diamond and Onwuegbuzie also used the ERAS to examine reading attitudes in a sample of 1,968 kindergarten kindergarten [Ger.,=garden of children], system of preschool education. Friedrich Froebel designed (1837) the kindergarten to provide an educational situation less formal than that of the elementary school but one in which children's creative play instincts would be  through fifth graders. They reported that females had higher scores on both reading attitudes and reading achievement measured by the Iowa Test of Basic Skills The Iowa Test of Basic Skills (ITBS) are a set of standardized tests given annually to school students in the United States. These tests are given to students beginning in kindergarten and progressing until Grade 8 to assess educational development. .

Kush et al. (1995) examined the attitudes of 139 males and 150 females attending grades 1 to 5 in a suburban school district in the southwestern United States United States, officially United States of America, republic (2005 est. pop. 295,734,000), 3,539,227 sq mi (9,166,598 sq km), North America. The United States is the world's third largest country in population and the fourth largest country in area. . They found significant gender differences in attitudes on all three subscales on two separate administrations of the ERAS in grades 1, 3, 4, and 5. There were no differences between genders at the second-grade level. Ley and Trentham (1987) found that gifted girls in grades 7 and 8 displayed significantly more positive attitudes toward reading than gifted boys.

The findings with regard to grade level have been a little less consistent. For example, Diamond and Onwuegbuzie (2001) found that reading attitudes declined substantially beginning in the fourth grade. McKenna et al. (1995) also noted that attitudes toward reading decreased as students progressed to higher grades, except for readers with high-ability scores; these readers rated themselves as having generally consistent positive attitudes toward recreational reading across grades. However, Kush et al. (1995) found no differences in reading attitudes among grade levels in a suburban sample of students in grades 1 to 5.

The Present Study

The research on attitudes toward reading has revealed several trends. Reading attitudes can be measured reliably at the elementary-school level (e.g., McKenna & Kear, 1990), higher-ability students have more positive attitudes toward reading than lower-ability students at the secondary level (e.g., Kennedy & Halinski, 1975; Ley & Trentham, 1987), reading attitude scores decrease as grade level increases except in higher-ability students (Diamond & Onwuegbuzie, 2001; McKenna et al., 1995), and girls have more positive attitudes toward reading than boys (e.g., Cloer & Dalton Dalton, city (1990 pop. 21,761), seat of Whitfield co., extreme NW Ga., in the Appalachian valley; inc. 1847. It is a highly industrialized city in a farm area. , 2001; Diamond & Onwuegbuzie; Kush et al., 1995).

In this study, the construct validity construct validity,
n the degree to which an experimentally-determined definition matches the theoretical definition.
 of the ERAS was investigated in a sample of academically talented (AT) elementary-aged students. First, the internal consistency of scores on the ERAS subscales as defined by McKenna and Kear (1990) were examined. Second, the percentile ranks The percentile rank of a score is the percentage of scores in its frequency distribution which are lower. For example, a test score which is greater than 85% of the scores of people taking the test is said to be at the 85th percentile.  of these academically talented participants were compared to the published norms for the ERAS. It was hypothesized that the AT students would have attitudes towards reading that were at the upper end of the published norms, which were based on a heterogeneous ability sample (McKenna & Kear). Third, gender differences were examined to see if AT girls and boys differed on reading attitudes. Although this finding has been reported on gifted students at the middle-school level (Ley & Trentham, 1987), the question has not been examined with elementary-aged students or with AT students. Fourth, grade level differences were examined to see if reading attitudes were lower in the upper grades for this high-achieving group. Finally, given the potential restriction of score ranges in a group of AT students, ERAS scores were examined to see if they had structural validity in this population.



The participants consisted of 575 students (277 females, 298 males) who attended a university-based summer program for AT youth in the western U.S. Students ranged in age from 5 to 12 (M age = 8.88, SD = 1.85), and had completed kindergarten through grade 6 the previous academic year. A crosstabulations analysis indicated that the number of students did not differ significantly across grade levels by gender, [chi square chi square (kī),
n a nonparametric statistic used with discrete data in the form of frequency count (nominal data) or percentages or proportions that can be reduced to frequencies.
] (6, N = 575) = 4.64, p > .05. No data were collected on ethnicity ethnicity Vox populi Racial status–ie, African American, Asian, Caucasian, Hispanic  of participants, but the ethnic breakdown for the entire program was 45% White, 17% Chinese American Chinese Americans (Chinese language: 美籍華人 or 華裔美國人) are Americans of Chinese descent. Chinese Americans constitute one group of Overseas Chinese and are a subgroup of Asian Americans. , 12% Mixed, 11% Other Asian American A·sian A·mer·i·can also A·sian-A·mer·i·can  
A U.S. citizen or resident of Asian descent. See Usage Note at Amerasian.

, 7% African American African American Multiculture A person having origins in any of the black racial groups of Africa. See Race. , 5% Chicano or Latino, and 4% not reported. Students in the program also came from a range of socioeconomic so·ci·o·ec·o·nom·ic  
Of or involving both social and economic factors.


of or involving economic and social factors

Adj. 1.
 groups; however, as in many programs for talented students, the majority of participants were at least middle to upper-middle class with a median family income in the $80,000 to $100,000 range.

The summer program is a 3-week, nonresidential program for AT youth. Students attend one class for 3.5 hours a day (either morning or afternoon) 4 days a week. Admission to the program is competitive and applicants to the program must submit report cards, a work sample, and a teacher recommendation. Standardized standardized

pertaining to data that have been submitted to standardization procedures.

standardized morbidity rate
see morbidity rate.

standardized mortality rate
see mortality rate.
 achievement test scores are also required for the upper-elementary grades. Students do not have to be identified as gifted in their home schools to qualify for the program. However, on average, 10% of the students in kindergarten through grade 2 and 35% of the students in grades 3 to 6 typically report that they have been identified as gifted and talented in their home schools. The majority of students in the primary grades (64%) do not have the option at their grade levels; and about 50% of the students do not attend schools with Gifted and Talented Education (GATE) programs. Course offerings come from the disciplines of mathematics, natural science, social science, and writing. The faculty consist of outstanding public- and private-school teachers and university-graduate students.


The ERAS (McKenna & Kear, 1990) is a 20-item instrument designed for students in grades 1 through 6. The first 10 items assess attitudes toward recreational reading (e.g., "How do you feel when you read a book on a rainy rain·y  
adj. rain·i·er, rain·i·est
Characterized by, full of, or bringing rain.

raini·ness n.

 Saturday?"), and the next 10 assess attitudes toward academic reading (e.g., "How do you feel about taking a reading test?"). The scale has a 4-point response system using pictorial anchors. Each of the four anchors shows a picture of the cartoon cartoon [Ital., cartone=paper], either of two types of drawings: in the fine arts, a preliminary sketch for a more complete work; in journalism, a humorous or satirical drawing.  cat, Garfield. The most positive picture shows Garfield with his hands in the air and a big grin on his face. The most negative figure shows Garfield scowling scowl  
v. scowled, scowl·ing, scowls

To wrinkle or contract the brow as an expression of anger or disapproval. See Synonyms at frown.
 with his arms tensed at his side and fists clenched clench  
tr.v. clenched, clench·ing, clench·es
1. To close tightly: clench one's teeth; clenched my fists in anger.

. The more positive of the two middle options shows Garfield smiling whereas Garfield's face is unhappy in the other picture. For each of the 20 items, students choose the Garfield that is most representative of their feelings. Scores on the two subscales range from 10 to 40 and scores on the total scale range from 20 to 80.

There is extensive psychometric information available on the measure. The internal consistency estimates (Cronbach, 1951) for scores in the six grades in the normative sample ranged from .74 to .89, with only two of the estimates falling below .80 (McKenna & Kear, 1990). McKenna and Kear (p. 639) established the two subscales using exploratory factor analysis, and reported that scores on both the recreational and academic subscales were moderately correlated (r = .64). ERAS scores have shown evidence of convergent validity (Allen et al., 1992), and Kush et al. (1995) reported that ERAS scores showed moderate stability (.29 [less than or equal to] r [less than or equal to] .43, Mdn = .39) over a 1-year period.


The ERAS was mailed in the spring of 1997 to parents of students who were accepted into the program for the summer of 1997 with their acceptance packets. Parents were told that they should assist their children in completing the survey, including reading the questions if necessary, but to allow children to choose their own responses. A pre-addressed, stamped, return envelope was provided and parents were told to put no identifying information on the questionnaire. Questionnaires were returned in the summer and all responses were anonymous. Six hundred and eight surveys were returned, representing 65% of the 1997 elementary-program cohort cohort /co·hort/ (ko´hort)
1. in epidemiology, a group of individuals sharing a common characteristic and observed over time in the group.

. Thirty-three out of the 608 (5.4%) were excluded due to missing data on grade, gender, or age. Missing data on individual reading items in the reduced sample of 575 ranged from zero (Item 1) to 19 (Item 16) participants, and these missing items were replaced with the mean score for that item. The biggest change in the standard deviation In statistics, the average amount a number varies from the average number in a series of numbers.

(statistics) standard deviation - (SD) A measure of the range of values in a set of numbers.
 when the missing values In statistics, missing values are a common occurrence. Several statistical methods have been developed to deal with this problem. Missing values mean that no data value is stored for the variable in the current observation.  were replaced was .02 on Item 16. The final sample of 575 constituted 62% of the potential participants.


Internal Consistency

Table 1 contains internal consistency estimates for scores on the three ERAS scales as defined by McKenna and Kear (1990) for each grade and gender. As the table shows, internal consistency scores were generally satisfactory. With the exception of the scores for students entering grade 1 and females entering grade 3, reliability estimates were above .70. Because the median is less affected by distribution asymmetry Asymmetry

A lack of equivalence between two things, such as the unequal tax treatment of interest expense and dividend payments.
 than the mean (Huck huck  

Noun 1. huck - toweling consisting of coarse absorbent cotton or linen fabric

toweling, towelling - any of various fabrics (linen or cotton) used to make towels
, 2000), the former was used to indicate central tendency for the reliability estimates. Median internal consistency coefficients for the recreational, academic, and total scores were .79, .84, and .87, respectively.

Comparison to Normative Sample

Participants' minimum, maximum, median, and mean raw scores were calculated by grade level and were then converted to percentiles using the ERAS tables from the normative sample (McKenna & Kear, 1990). As the data for the normative sample were collected in January and the data in this study were collected in the summer, participants' percentiles were calculated based on the norm of the grade they would be entering in the fall semester se·mes·ter  
One of two divisions of 15 to 18 weeks each of an academic year.

[German, from Latin (cursus) s
, rather than the grade that they had just completed. However, students who had completed grade 6 were compared to the grade 6 norms; there are no grade 7 norms. Table 2 contains the percentile ranks by grade level for each of the three scores (academic, recreational, and total). As the table indicates, median percentile ranks did not fall below the 70th percentile percentile,
n the number in a frequency distribution below which a certain percentage of fees will fall. E.g., the ninetieth percentile is the number that divides the distribution of fees into the lower 90% and the upper 10%, or that fee level
 for any grade level, and the lowest mean percentile was the 67th. The highest scores at all grade levels were at the 99th percentile. These data indicate that the distribution of scores for this sample is at the upper end of the distribution of scores for the normative sample.

Gender and Grade-Level Differences

As the academic and recreational scores were moderately correlated (r = .62), differences on recreational and academic reading scores were examined using a two-factor multivariate analysis multivariate analysis,
n a statistical approach used to evaluate multiple variables.

multivariate analysis,
n a set of techniques used when variation in several variables has to be studied simultaneously.
 of variance (MANOVA MANOVA Multivariate Analysis of the Variance ), with grade and gender as the two factors (see Table 3). The results of the MANOVA (Pillai's Trace) indicated significant main effects for grade, F(12, 1122) = 4.65, p = .001, multivariate The use of multiple variables in a forecasting model.  effect size = .05; and gender, F(2,560) = 14.32, p = .001, multivariate effect size = .05; but no interaction effect, F(12, 1122) = .90, p = .545, multivariate effect size = .01. These findings need to be interpreted with caution as the sample sizes were unequal and the tests of homogeneity Homogeneity

The degree to which items are similar.
 of variance were significant.

Descriptive discriminant function discriminant function
n. Statistics
A function of a set of variables used to classify an object or event.
 analyses (DFA DFA - Deterministic Finite-state Automaton. See Finite State Machine. ), the recommended post-hoc procedure for MANOVA (Weinfurt, 1995), were calculated by grade and by gender. The DFA by grade resulted in one significant function, Wilks' lambda = .92, [chi square] (12) = 52.1, p < .001. However, function classified only 24.9% of the students correctly, suggesting that significance was more a function of sample size than of effect size, and no further grade-level analyses were conducted. Academic reading ([beta] = 1.23, s = .98) was more important in this function than recreational reading ([beta] = -.23, s = .46). The DFA by gender was also significant, Wilks' lambda = .93, [chi square] (2) = 44.90, p < .001. In this analysis, 61% of the sample was correctly classified, and recreational reading scores ([beta] = .85, s = .98) proved to be more important in classifying students than academic reading scores ([beta] = .22, s = .73).

Post-hoc gender comparisons were conducted at each grade level on the two sets of scores. The alpha level was set at .004 for each comparison to control for Type I error using the Bonferroni correction In statistics, the Bonferroni correction states that if an experimenter is testing n independent hypotheses on a set of data, then the statistical significance level that should be used for each hypothesis separately is 1/n . Cohen's d was used as an effect size measure with values in the .5 range indicating a medium effect size, and values in the .8 range indicating a large effect size (Newton & Rudestam, 1999). Results indicated that females entering grades 4 (Cohen's d = .65), 5 (Cohen's d = .56), and 6 (Cohen's d = .88) obtained significantly higher scores than males on Recreational Reading, and that females entering grade 6 (Cohen's d = .89) obtained significantly higher scores than males on Academic Reading. Effect sizes also indicated substantial differences in favor of upon the side of; favorable to; for the advantage of.

See also: favor
 females entering grade 3 (Cohen's d = .88) and grade 7 (Cohen's d = .49) on Recreational Reading, but these differences did not achieve the critical alpha for the comparisons. No other gender differences were significant at the critical alpha level, nor did they have medium or large effect sizes.

Factor Structure

The factor structure was examined using scores from all participants. Since there has been only one examination of the instrument's factor structure (i.e., the normative study, McKenna & Kear, 1990), and the earlier comparison of this sample's scores to the normative sample indicated a restriction in the range of this sample's scores, exploratory factor analysis was used. In keeping with McKenna and Kear, an unweighted least squares extraction was performed with varimax rotation. Bartlett's Test Bartlett's test (Snedecor and Cochran, 1983) is used to test if k samples have equal variances. Equal variances across samples is called homoscedasticity or homogeneity of variances.  of Sphericity was significant (BTS BTS - Bug Tracking System  value = 3688.58, p < .001) and the KMO KMO Kaiser-Meyer-Olkin (test to assess the appropriateness of using factor analysis on data)
KMO Knowledge Master Open (academic competition)
KMO Kunglig Majestäts Orden
 Measure of Sampling Adequacy was substantial (r = .92). Four factors with eigenvalues eigenvalues

statistical term meaning latent root.
 greater than 1 were extracted; however, the scree test and parallel analysis using the tables provided by Lautenschlager (1989) and a computer program by Watkins (2000) indicated the presence of two factors: low communality estimates (ranging from .21 to .48) and a variable/factor ratio of 10 to 1 indicated that a sample size of at least 200 was adequate for a stable factor solution (MacCallum, Widaman, Zhang, & Hong, 1999). Item salience sa·li·ence   also sa·li·en·cy
n. pl. sa·li·en·ces also sa·li·en·cies
1. The quality or condition of being salient.

2. A pronounced feature or part; a highlight.

Noun 1.
 was set at .35, indicating at least 12% shared variance between the item and the factor.

Two-, three-, and four-factor solutions were examined. The results of the two-factor solution, including the communality estimates, are presented in Table 4. The two factors accounted for 36% of the variance and were similar to the two factors reported by McKenna and Kear (1990). Factor I, labeled Academic Reading, consisted of 12 items with factor loadings at or above .35: the 10 designated academic reading items and 2 recreational reading items. Factor II, labeled Recreational Reading, consisted of 9 of the 10 recreational reading items and one academic reading item that cross-loaded at .41. Construct reliability estimates based on the items with salient loadings (i.e., at least .35) were .82 and .79 for Factors I and II, respectively.

The three-factor solution, which accounted for 39% of the variance, yielded three interpretable factors based on factor loadings of .35 or above. Factor I was labeled Reading for Fun, and consisted of eight items from the recreational subscale, These items focused on reading as a pleasurable pleas·ur·a·ble  
Agreeable; gratifying.

 activity and an alternative to other activities like playing. Factor II, labeled Reading for Learning, consisted of eight items focusing on reading activities specifically related to learning (e.g., taking a reading test, reading workbook work·book  
1. A booklet containing problems and exercises that a student may work directly on the pages.

2. A manual containing operating instructions, as for an appliance or machine.

 pages). The third factor was made up of five items and was labeled Reading in School. All five items focused on reading that takes place in school, but without the learning component (e.g., reading in school during free time or reading out loud in class). In this solution, three items cross-loaded on two factors and two items (9 and 10) did not achieve a salient loading on any factor. Construct reliability estimates for the three factors' scores were .76, .75, and .66, respectively.

The four-factor solution accounted for 41% of the variance and also produced interpretable factors. The first three factors were similar to those from the three-factor solution and were labeled Reading for Learning (8 items), Reading for Fun (6 items), and Reading in School (5 items), respectively. The fourth factor consisted of three items--the two that did not have salient loadings on any factor in the three-factor solution, and an item that had originally loaded on the Reading for Fun factor in the three-factor solution. This factor, consisting of items related to books (e.g., going to a bookstore, starting a new book, reading different kinds of books), was labeled Reading Books. The construct reliability estimates of scores on the four factors based on salient loadings were in the moderate to low range--.75, .74, .61, and .55, respectively. Moreover, item four did not achieve a salient loading on any factor, and three items had salient loadings on two factors.

In summary, the two-factor structure bad one less cross-loading item than the three- and four-factor structures; it did not differ substantially on the basis of variance accounted for; and it had construct reliability estimates in the .8 range for the subscale scores, estimates that were lower for the three- and four-factor subscale scores. Additionally, all 20-ERAS items obtained salient loadings on the two-factor solution. On the basis of these results, previous findings, and the theoretical underpinning un·der·pin·ning  
1. Material or masonry used to support a structure, such as a wall.

2. A support or foundation. Often used in the plural.

3. Informal The human legs. Often used in the plural.
 of the instrument (McKenna & Kear, 1990), the two-factor structure was accepted.


This study focused on the reliability and structural validity of scores on the ERAS in a sample of academically talented (AT) students. The ERAS' academic, recreational, and total scores were found to have satisfactory internal consistency across the elementary grades, and the distribution of scores of AT students was above average when compared to the ERAS normative sample. Gender differences were found on reading attitudes at some grade levels; however, no meaningful grade-level differences were found. Further, although factor analysis supported the two-factor structure of ERAS scores reported by McKenna and Kear (1990), the analyses also suggested that academic and recreational reading attitudes may both be composed of multiple interrelated in·ter·re·late  
tr. & intr.v. in·ter·re·lat·ed, in·ter·re·lat·ing, in·ter·re·lates
To place in or come into mutual relationship.


Internal Consistency of ERAS Scores

The internal consistency of ERAS subscale scores (see Table 1) indicates that the instrument produces reliable information across the elementary-school years. Forty-eight internal consistency estimates were calculated on scores by grade and gender with samples ranging from 9 to 298, and 45 (94%) of the estimates were .72 or higher. Moreover, all but one of the estimates based on the 20-item total score were above .80. Only scores on the recreational subscale with second-grade girls and scores for the small first-grade samples did not achieve an adequate level of consistency. Allen et al. (1992) reported split-half reliability coefficients on scores of 46-fifth graders of .88 for the recreational scale and .81 for the academic subscale. In the context of previous findings and the other results in this study, the few low reliability coefficients are probably anomalous a·nom·a·lous  
1. Deviating from the normal or common order, form, or rule.

2. Equivocal, as in classification or nature.
, or due to the low sample size in the first-grade-gender cohorts.

Comparison of AT Students to Published Norms

When the scores of this sample's participants were converted to percentiles based on the ERAS normative sample, the findings clearly indicate above average performance of the group as a whole. Although the maximum percentile is the 99th for scores at all grade levels, the more critical information is found in the measures of central tendency. The median value Noun 1. median value - the value below which 50% of the cases fall

statistics - a branch of applied mathematics concerned with the collection and interpretation of quantitative data and the use of probability theory to estimate population
 of the means calculated in this study is at the 77.5th percentile, and the median values of the median and mode are at the 79.5th and the 85.5th percentiles, respectively. At the same time, some participants were clearly at the lower end of the distribution (see minimum scores in Table 2). It is not clear if the students with lower scores represent individuals with lower achievement who were selected on the basis of teacher recommendations, or individuals who may be less able to benefit from a gifted program.

It is possible that students with unfavorable attitudes toward reading can be influenced by participation in a community where many individuals have very positive attitudes toward reading. These are questions that should be addressed in a further study. If reading attitudes have a positive impact on achievement and being around students with positive attitudes results in a substantial increase in other's attitudes, putting students into academic communities can be another way of facilitating academic achievement, and may support claims for heterogeneous grouping in schools. However, if positive attitudes toward reading are transmitted through interaction, students in relatively homogenous homogenous - homogeneous  groups such as gifted and talented programs may benefit from being around others with similar attitudes. One goal of future research may be to find the right balance between students with more and less positive attitudes, so that the transmission of attitudes is in the pro-reading direction for all students.

Grade and Gender Differences on ERAS Scores

Significant differences between males and females were found for 4 of the 14 gender comparisons made in this study with medium to large effect sizes; and consistent with previous research literature, on all of the comparisons, females obtained higher scores. Moreover, all but one of the meaningful differences was in recreational reading. Kush et al. (1995) reported significant gender differences favoring favoring

an animal is said to be favoring a leg when it avoids putting all of its weight on the limb. A part of being lame in a limb.
 females on ERAS recreational-reading scores at grades l, 3, 4, and 5; and McKenna et al. (1995) also reported higher female scores on the ERAS. McKenna et al. hypothesized that unexplained unexplained

strange or unclear because the reason for it is not known

Adj. 1. unexplained - not explained; "accomplished by some unexplained process"
 cultural expectations may contribute to girls' more positive attitudes toward reading. Like Kush et at., no meaningful grade-level differences were found in this study, nor was there a decline in reading scores across the grades. Thus, these findings also replicate rep·li·cate
1. To duplicate, copy, reproduce, or repeat.

2. To reproduce or make an exact copy or copies of genetic material, a cell, or an organism.

A repetition of an experiment or a procedure.
 McKenna et al.'s findings for high-ability students.

McKenna et al. (1995) also found that students rated by their teachers as stronger readers had significantly more positive recreational reading attitudes, but their academic reading attitudes were no higher than students who were rated as average and poor readers. The current findings suggest that even in AT populations, girls are likely to have more positive attitudes toward recreational reading than boys. It is not clear if positive attitudes toward reading leads to greater reading achievement or vice versa VICE VERSA. On the contrary; on opposite sides. , but this question should be investigated empirically.

Factor Structure of ERAS Scores

The current results on the ERAS factor structure essentially replicated McKenna and Kear's (1990) findings: Two factors were extracted which accounted for 36% of the variance. They were labeled Academic and Recreational Reading and the interfactor correlation was similar to the one reported by McKenna and Kear. However, further analyses conducted to reduce the risk of underextraction of factors revealed two additional factors. The Academic Reading factor split into a Reading for Learning component and a Reading in School component, and the Recreational Reading factor split into Reading for Fun and Reading Books factors. While these sub-factors were not viable (e.g., low internal consistency estimates), the items formed a recognizable cluster on the basis of interitem correlations and on the basis of content, Additionally, their emergence reminds us that academic and recreational reading attitudes are made up of subcomponents and that a valid measure of reading attitudes needs to tap into different types of attitudes about reading.


There were several limitations in the current study. First, there were no measures of either ethnicity or socioeconomic status socioeconomic status,
n the position of an individual on a socio-economic scale that measures such factors as education, income, type of occupation, place of residence, and in some populations, ethnicity and religion.
; thus, it was not possible to see if there would be differences among groups on these variables. Some researchers have reported ethnic group differences in reading achievement (e.g., Diamond & Onwuegbuzie, 2001), although these achievement differences have not been demonstrated in research on attitudes, and the studies on differences in attitudes toward reading among socioeconomic groups have yielded mixed results (see Wigfield & Asher, 2000, for a summary of the literature). Second, there were no measures of convergent or divergent di·ver·gent  
1. Drawing apart from a common point; diverging.

2. Departing from convention.

3. Differing from another: a divergent opinion.

 validity. Future studies of the ERAS in academically talented populations should examine the concurrent and predictive validity In psychometrics, predictive validity is the extent to which a scale predicts scores on some criterion measure.

For example, the validity of a cognitive test for job performance is the correlation between test scores and, for example, supervisor performance ratings.
 of the instrument's subscale scores.


The results obtained in this study support the structural validity of the ERAS (McKenna & Kear, 1990), and contribute to the body of construct validity evidence (e.g., Allen et al., 1992; Kush et at., 1995; McKenna & Kear; McKenna et at., 1995) for the instrument's scores. They also raise some interesting questions for future research. The internal consistency of scores on the two factors reported by McKenna and Kear are substantial for a measure of attitudes in elementary-aged populations. The support for the reliability of ERAS scores in an AT sample extends the generalizability of the instrument, and the study has shown that groups chosen for high achievement do have more positive attitudes toward reading than a more representative sample of the elementary-aged school population.

The gender differences that have been noted in previous studies of attitudes toward reading were replicated in this study, suggesting that these differences are pervasive and may be found in students at all achievement levels. Perhaps the most interesting finding, however, is that both Academic and Recreational Reading Attitudes may be more complex than previously determined. Thus, we echo Wigfield and Asher's (2000) call for investigations into "specific dimensions of reading attitudes and motivation to read rather than ... [more examinations of] the global 'attitude toward reading' construct" (p. 434).

In conclusion, the ERAS is yet another instrument that can be used in working and research with AT students, including those who are not living up to their academic potential. Attitudes toward reading, particularly recreational reading, may prove to be useful markers in educational programs for elementary-aged AT populations.

Manuscript submitted July 7, 2004.

Revision accepted October 5, 2004.


Allen, L., Cipielewski, J., & Stanovich, K. E. (1992). Multiple indicators of children's reading habits and attitudes: Construct validity and cognitive correlates. Journal of Educational Psychology, 84, 489-503.

Anderson, M. A., Tollefson, N. A., & Gilbert, E. C. (1985). Giftedness gift·ed  
1. Endowed with great natural ability, intelligence, or talent: a gifted child; a gifted pianist.

 and reading: A cross sectional sec·tion·al  
1. Of, relating to, or characteristic of a particular district.

2. Composed of or divided into component sections.

 view of differences in reading attitudes and behaviors. Gifted Child gifted child

Child naturally endowed with a high degree of general mental ability or extraordinary ability in a specific domain. Although the designation of giftedness is largely a matter of administrative convenience, the best indications of giftedness are often those
 Quarterly, 29, 186-189.

Askov, E., & Fishback, T. (1973). An investigation of primary pupils' attitudes toward reading. Journal of Experimental Education, 41, 1-7.

Cloer, T., Jr., & Dalton, S. R. (2001). Gender and grade differences in reading achievement and in self-concept as readers. Journal of Reading Education, 26(2), 31-36.

Cooter, R. B., Jr., & Alexander, J. E. (1984). Interest and attitude: Affective affective /af·fec·tive/ (ah-fek´tiv) pertaining to affect.

1. Concerned with or arousing feelings or emotions; emotional.

 connections for gifted and talented readers. Reading World, 24, 97-102.

Cronbach, L. J. (1951). Coefficient coefficient /co·ef·fi·cient/ (ko?ah-fish´int)
1. an expression of the change or effect produced by variation in certain factors, or of the ratio between two different quantities.

 alpha and the internal structure of tests. Psychometrika, 16, 434-439.

Diamond, P. J., & Onwuegbuzie, A. J. (2001). Factors associated with reading achievement and attitudes among elementary school-aged students. Research in the Schools, 8, 1-11.

Huck, S. W. (2000). Reading statisties and research (3rd ed.). Menlo Park Menlo Park.

1 Residential city (1990 pop. 28,040), San Mateo co., W Calif.; inc. 1874. Electronic equipment and aerospace products are manufactured in the city. Menlo College and a Stanford Univ. research institute are there.

2 Uninc.
, CA: Longman.

Kennedy, L. D., & Halinski, R. S. (1975). Measuring attitudes: An extra dimension. Journal of Reading, 18, 518-522.

Kush, J. C., Watkins, M. W., McAleer, A. T., & Edwards, V. A. (1995). One-year stability of the Elementary Reading Attitude Survey. Mid-Western Educational Researcher, 8(1), 11-14.

Lautenschlager, G. J. (1989). A comparison of alternatives to conducting Monte Carlo Monte Carlo (môNtā` kärlō`), town (1982 pop. 13,150), principality of Monaco, on the Mediterranean Sea and the French Riviera.  analyses for determining parallel analysis criteria. Multivariate Behavioral Research, 24, 365-395.

Ley, T. C., & Trentham, L. L. (1987). The reading attitudes of gifted learners in grades seven and eight. Journal for the Education for the Gifted, 10, 87-98.

MacCallum, R. C., Widaman, K. F., Zhang, S., & Hong, S. (1999). Sample size in factor analysis. Psychological Methods, 4, 84-99.

McKenna, M. C, & Kear, D. J. (1990). Measuring attitude toward reading: A new tool for teachers. The Reading Teacher, 43, 626-639.

McKenna, M. C., Kear, D. J., & Ellsworth, R. A. (1995). Children's attitudes toward reading: A national survey. Reading Research Quarterly, 30, 934-956.

Mikulecky, L. J. (1976). The developing, field testing, and initial norming of a secondary/adult level reading attitude measure that is behaviorally oriented o·ri·ent  
1. Orient The countries of Asia, especially of eastern Asia.

a. The luster characteristic of a pearl of high quality.

b. A pearl having exceptional luster.

 and based on Krathwohl's taxonomy taxonomy: see classification.

In biology, the classification of organisms into a hierarchy of groupings, from the general to the particular, that reflect evolutionary and usually morphological relationships: kingdom, phylum, class, order,
 of the affective domain affective domain,
n the area of learning involved in appreciation, interests, and attitudes.
 (Doctoral dissertation dis·ser·ta·tion  
A lengthy, formal treatise, especially one written by a candidate for the doctoral degree at a university; a thesis.


, University of Wisconsin-Madison “University of Wisconsin” redirects here. For other uses, see University of Wisconsin (disambiguation).
A public, land-grant institution, UW-Madison offers a wide spectrum of liberal arts studies, professional programs, and student activities.
, 1976). Ann Arbor Ann Arbor, city (1990 pop. 109,592), seat of Washtenaw co., S Mich., on the Huron River; inc. 1851. It is a research and educational center, with a large number of government and industrial research and development firms, many in high-technology fields such as , MI: University Microfilms, Order No. 76-28, 162. (ERIC Document Reproduction Service No. ED146550)

Newton, R. R., & Rudestam, K. E. (1999). Your statistical consultant: Answers to your data analysis questions. Thousand Oaks Thousand Oaks, residential city (1990 pop. 104,352), Ventura co., S Calif., in a farm area; inc. 1964. Avocados, citrus, vegetables, strawberries, and nursery products are grown. , CA: Sage.

Ross, E. P., & Fletcher, R. K. (1989). Responses to children's literature children's literature, writing whose primary audience is children.

See also children's book illustration. The Beginnings of Children's Literature

The earliest of what came to be regarded as children's literature was first meant for adults.
 by environment, grade level, and sex. Reading Instruction Journal 32, 22-28.

Smith, M. C. (1990). A longitudinal lon·gi·tu·di·nal
Running in the direction of the long axis of the body or any of its parts.
 investigation of reading attitude development from childhood to adulthood. Journal of Educational Research, 83, 215-219.

Swanson, B. B. (1982). The relationship between attitudes toward reading and reading achievement. Educational and Psychological Measurement, 42, 1303-1304.

Walberg, H. J., & Tsai, S. L. (1985). Correlates of reading achievement and attitude: A national assessment study. Journal of Edacational Research. 78, 159-167.

Watkins, M. W. (2000). MacParallel analysis [Computer software]. State College, PA: Ed & Psych psych also psyche   Informal
v. psyched, psych·ing, psyches
a. To put into the right psychological frame of mind:

Weinfurt, K. P. (1995). Multivariate analysis of variance. In L. G. Grimm & P. R. Yarnold (Eds.), Reading and understanding multivariate statistics Multivariate statistics or multivariate statistical analysis in statistics describes a collection of procedures which involve observation and analysis of more than one statistical variable at a time. Sometimes a distinction is made between univariate (e.g.  (pp. 245-276). Washington, DC: American Psychological Association The American Psychological Association (APA) is a professional organization representing psychology in the US. Description and history
The association has around 150,000 members and an annual budget of around $70m.

Wigfield, A., & Asher, S. R. (2000). Social and motivational influences on reading. In M. L. Kamil, P. B. Mosenthal. P. D. Pearson, & R. Barr (Eds.), Handbook of reading research. Vol. 3. (pp. 423-452). Mahwah, NJ: Lawrence Erlbaum.

This research was conducted under the auspices aus·pi·ces 1  
Plural of auspex.

Noun, pl

under the auspices of with the support and approval of [Latin auspicium augury from birds]

 of the Academic Talent Development Program, University of California, Berkeley The University of California, Berkeley is a public research university located in Berkeley, California, United States. Commonly referred to as UC Berkeley, Berkeley and Cal .

Correspondence concerning this article should be addressed to Frank C. Worrell, PhD, Cognition cognition

Act or process of knowing. Cognition includes every mental process that may be described as an experience of knowing (including perceiving, recognizing, conceiving, and reasoning), as distinguished from an experience of feeling or of willing.
 and Development, 4511 Tolman Hall, Berkeley, CA 94720-1670. E-mail:; Telephone: (510) 643-4891; Fax: (510) 642-3555.

Frank C. Worrell directs the School Psychology program at the University of California, Berkeley and is Faculty Director of the Academic Talent Development Program. His research interests include academic talent development, at-risk youth, scale development and validation See validate.

validation - The stage in the software life-cycle at the end of the development process where software is evaluated to ensure that it complies with the requirements.
, and teacher effectiveness. He is a member of the editorial boards of Gifted Child Quarterly, Gifted Dialogue, the Journal of Advanced Academics The Journal of Advanced Academics is an academic journal that focuses on research that supports and enhances advanced academic achievement for students of all ages. In particular, JAA , Roeper Review, and School Psychology Quarterly, E-mail:

David A. Roth is Director of Special Education for the Piedmont Unified School District The Piedmont Unified School District (PUSD) is comprised of the seven schools in the city of Piedmont, California. Schools
  • Egbert W. Beach Elementary School
  • Frank C.
. His research interests include conduct problems and children's self-understanding. Recent work includes "Self-understanding and Academic Achievement: The Damon and Hart Model with High Achieving Urban School Students," with coauthors N. H. Gabelko, H. Asher, and E C. Worrell, presented at the American Educational Research Association's Annual Convention (2000). E-mail: droth@piedmont Piedmont, region, Italy
Piedmont (pēd`mŏnt), Ital. Piemonte, region (1991 pop. 4,302,565), 9,807 sq mi (25,400 sq km), NW Italy, bordering on France in the west and on Switzerland in the north.

Nina H. Gabelko is Director of the Academic Talent Development Program at the University of California, Berkeley. Her research interests are teaching gifted students, K-12; teacher education; professional growth of teachers; and social studies education. She is Editor of Cornerstones of Collaboration (1998), a collection of essays on the potential of partnerships to transform education, and author of Reducing Adolescent ad·o·les·cent
Of, relating to, or undergoing adolescence.

A young person who has undergone puberty but who has not reached full maturity; a teenager.
 Prejudice: A Handbook (1981), with John U. Michaelis; and Someone Just Like Me: When Academic Engagement Trumps trump 1  
1. Games
a. A suit in card games that outranks all other suits for the duration of a hand. Often used in the plural.

b. A card of such a suit.

c. A trump card.

 Race, Class, and Gender, with L. A. Sosniak in Phi Delta Kappan (2002). E-mail:
Table 1
Internal Consistency Estimates
for ERAS Scores by Grade and Gender


                  N    Recreational   Academic   Total

Entering 1 (a)    11       .64          .79       .85
Entering 2        29      79.00         .76       .85
Entering 3        39       .52          .75       .76
Entering 4        45       .83          .88       .89
Entering 5        47       .75          .84       .87
Entering 6        49      77.00         .87       .89
Entering 7        57       .87          .79       .89
All grades       277       .78          .84       .87


                  N    Recreational   Academic   Total

Entering 1 (a)    9        .26          .87       .81
Entering 2       32        .82          .87       .91
Entering 3       43        .86          .85       .91
Entering 4       49        .81          .79       .87
Entering 5       64        .73          .72       .81
Entering 6       37        .82          .80       .88
Entering 7       64        .89          .88       .91
All grades       298       .84          .84       .89

Note. ERAS = Elementary Reading Attitude Survey.

(a) Unstable due to small sample size.

Table 2
Percentile Ranks for ERAS
Raw Scores by Grade Level

Grade               N    Minimum   Median   Mean   Maximum

  Entering 1        20     38        77      77      99
  Entering 2        61     <1        79      74      99
  Entering 3        82      3        81      75      99
  Entering 4        94      1        84      78      99
  Entering 5       111     16        87      82      99
  Entering 6        86     15        88      83      99
  Entering 7 (a)   121      1        79      74      99
  Entering 1        20      5        75      69      99
  Entering 2        61      1        73      67      99
  Entering 3        82     <1        74      74      99
  Entering 4        94      3        86      75      99
  Entering 5       111     17        82      82      99
  Entering 6        86     <1        82      82      99
  Entering 71      121      3        73      73      99
  Entering 1        20     20        75      72      99
  Entering 2        61     <1        74      71      99
  Entering 3        82     <1        79      73      99
  Entering 4        94      4        80      78      99
  Entering 5       111     36        87      87      99
  Entering 6        86      3        86      84      99
  Entering 71      121      1        76      76      99

Note. ERAS = Elementary Reading Attitude Survey.

(a) Based on grade 6 percentile norms.

Table 3
Means and Standard Deviations from
ERAS Raw Scores by Grade and Gender

                    Recreational      Academic         Total

               N      M       SD      M       SD      M       SD
  Entering 1   11   34.82    4.02   33.90    4.69   68.72    8.29
  Entering 2   29   35.28    4.27   33.24    4.97   68.51    8.16
  Entering 3   39   35.12    3.33   33.23    4.92   68.35    7.07
  Entering 4   45   35.49    4.00   32.83    5.64   68.32    8.49
  Entering 5   47   35.51    3.63   31.80    5.19   67.32    7.86
  Entering 6   49   35.53    3.62   31.71    6.14   67.24    8.73
  Entering 7   57   33.66    4.72   28.84    4.78   62.50    8.50
  Entering 1    9   35.22    2.82   32.85    6.87   68.07    8.39
  Entering 2   32   33.04    5.31   30.41    6.54   63.45   11.08
  Entering 3   43   32.19    5.92   30.65    6.05   62.84   11.05
  Entering 4   49   32.59    4.91   30.02    5.55   62.61    9.61
  Entering 5   64   33.41    3.86   30.91    4.16   64.32    7.01
  Entering 6   37   31.88    4.80   26.62    5.18   58.50    9.01
  Entering 7   64   30.96    6.22   27.05    6.75   58.01    11.4

Note. ERAS = Elementary Reading Attitude Survey.

(a) Males and females differ at the .004 level. (b) Effect sizes for
gender based on Cohen's dare in the medium range (.49 - .65). (c)
Effect sizes for gender based on Cohen's dare in the large range
(.88 - .89).

Table 4
Two-Factor Structure of the
Elementary Reading Attitude Survey (N = 575)

                                  Factor I    Factor II     [h.sup.2]
ERAS items                        Academic   Recreational

  How do you feel:
    14. about reading your
        school books?               .67          .18           .44
    11. when the teacher asks
        you questions about
        what you read?              .66          .19           .44
    12. about doing reading
        workbook pages and
        worksheets?                 .65          .10           .41
    16. when it's time for
        reading class?              .57          .31           .43
    20. about taking a reading
        test?                       .54          .26           .38
    15. about learning from a
        book?                       .53          .23           .33
    17. about the stories you
        read in reading class?      .53          .17           .37
    18. when you read out loud
        in class?                   .50          .24           .31
    13. about reading in
        school?                     .47          .41           .41
    19. about using a
        dictionary?                 .43          .14           .21
    10. about reading different
        types of books? (a)         .41          .33           .30
     7. about reading during
        summer vacation?            .22          .68           .48
     5. about spending free
        time reading?               .24          .67           .48
     3. about reading for fun
        at home?                    .04          .65           .37
     8. about reading instead
        of playing?                 .33          .59           .47
     1. when you read a book on
        a rainy Saturday?           .14          .56           .33
     2. when you read a book in
        school during free
        time?                       .33          .46           .38
     6. about starting a new
        book?                       .30          .40           .28
     4. about getting a book
        for a present? (b)          .35          .38           .29
     9. about going to a
        bookstore?                  .21          .35           .22
Eigenvalues                        6.07         1.15
% Variance accounted for          30.3          5.8
Construct reliability               .82          .79

Note. (a) Recreational reading item that loaded on academic reading
factor. (b) Recreational reading item that loaded on both academic
and recreational reading factors.
COPYRIGHT 2007 The Roeper School
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2007, Gale Group. All rights reserved. Gale Group is a Thomson Corporation Company.

 Reader Opinion




Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:Enhancing Literacy; Elementary Reading Attitude Survey
Author:Gabelko, Nina H.
Publication:Roeper Review
Geographic Code:1USA
Date:Jan 1, 2007
Previous Article:Development of an identification procedure for a large urban school corporation: identifying culturally diverse and academically gifted elementary...
Next Article:Constructing a synthetic-analytic framework for precocious writing.

Related Articles
Learning style difference between gifted and nongifted sixth and seventh grade students. (The scholarship of teaching and learning).
Protecting you/protecting me: evaluation of a student-led alcohol prevention and traffic safety program for elementary students.
Engaged students, involved families, and dramatically improved scores: LeapFrog SchoolHouse [TM] Literacy Center Intervention: Pinewood Elementary,...
Las Vegas District's literacy initiative improves students' academic chances: fastest-growing district taps LeapFrog SchoolHouse for portable,...
A Comparison of Young Children's Technology-enhanced and Traditional Responses to Texts: An Action Research Project.
Struggling readers: skilled readers emerge where reading in the content areas is taught and practiced.
From the editor's desk.

Terms of use | Copyright © 2014 Farlex, Inc. | Feedback | For webmasters