Motivation, Needs Support, and Language Arts Classroom Practices: Creation and Validation of a Measure of Young Adolescents’ Perceptions

Abstract

Early adolescence is a critical time for examining academic motivation, specifically motivation to read. To support self-determined motivation to read, students’ needs for competence, autonomy, and relatedness must be met within the classroom context. Because classroom instructional practices are a key component of adolescents’ daily experiences in the classroom, research that investigates the influence of these practices on students’ intrinsic motivation to read is needed. In addition, the perceptions of students regarding the degree to which classroom instructional practices meet students’ needs as well as the influence of classroom instructional practices on students’ motivation to read should be considered. The field is lacking an established measure of early adolescents’ perceptions of classroom instructional practices and the degree to which they support students’ needs (i.e., competence, autonomy) and intrinsic motivation to read. This study, guided by self-determination theory, sought to address this gap in the literature by developing and validating a measure called the Language Arts Reading Practices Survey (LARPS). This measure assessed student perceptions of the degree to which classroom instructional practices in the language arts classroom support students’ needs for competence, autonomy, and students’ self-determined motivation to read. The results of this study provide preliminary support for the validity of the LARPS.

Keywords:

• reading motivation
• literacy
• measurement
• instructional practices

Introduction

Early adolescence is a pivotal developmental period during which researchers investigate academic motivation, which typically declines during the middle grades (Eccles & Roeser, 2011; Maulana, Opdenakker, Stroet, & Bosker, 2013). Similar to trends in general academic motivation, research indicates a decline in students’ motivation to read during the middle grades. For example, eighth-grade students report lower levels of intrinsic motivation to read compared to sixth-grade students (Kelley & Decker, 2009; Lepper, Corpus, & Iyengar, 2005). These findings are important, as motivation to read is a key precursor to literacy outcomes including time spent reading (De Naeghel, Van Keer, Vansteenkiste, & Rosseel, 2012) and reading achievement (Guthrie, Klauda, & Ho, 2013; Kelley & Decker, 2009). It is currently unclear what factors contribute to decreases in adolescents’ motivation (Varuzza, Sinatra, Eschenauer, & Blake, 2014).

Contemporary research has tended to use general reading motivation measures that ignore specific contextual variables (e.g., Paige, 2011; Wigfield & Guthrie, 1997). These trends in reading motivation research have resulted in a simplified view of adolescents’ motivation to read (Schiefele, Schaffner, Moller, & Wigfield, 2012). Researchers have acknowledged that motivation to read may vary based on contextual factors (Miller & Faircloth, 2014) and academic discipline (Guay et al., 2010). In addition, researchers have indicated relationships between domain-specific motivation, such as motivation to read, and specific contextual factors that may include classroom instructional practices such as collaborative projects (Aarnoutse & Schellings, 2003; Guthrie & Humenick, 2004). Thus, researchers must consider contextual differences in adolescent reading motivation (Neugebauer, 2014). Few researchers have investigated the factors within specific academic contexts that may support or hinder adolescents’ motivation to read. In addition, few have investigated the degree to which young adolescents’ needs are met in the language arts (LA) classroom.

Classroom instructional practices, particularly those drawn from the adopted curriculum, often serve as a core structure within the classroom (Ball & Cohen, 1996; Delaney, Pitcher, Gillis, & Walker, 2014; Grossman & Thompson, 2008). Much of the research investigating the influence of classroom instructional practices has related to specific intervention programs, such as Concept Oriented Reading Instruction (CORI; Guthrie et al., 2007). Researchers have not explored the influence of specific classroom instructional practices in middle grades LA classrooms (e.g., graphic organizers, collaborative projects) on young adolescents’ needs and reading motivation. Examining the influence of such classroom instructional practices may provide insight into which specific literacy practices are supportive of students’ needs and promote their motivation to read.

Theoretical Framework

This study used self-determination theory (SDT) as the guiding theoretical framework for understanding motivation. Informed by SDT, motivation exists on a continuum from the least self-determined (amotivation) to the most self-determined (intrinsic motivation), with various levels of extrinsic motivation (external, introjected, identified, and integrated regulation) in between these two (Ryan & Deci, 2000, 2009). Social environments, such as the classroom, can greatly influence motivation (Ryan & Deci, 2009). Informed by SDT, adolescents’ motivation can be supported through the fulfillment of their needs for competence (positive perception of ability to successfully meet challenges and achieve mastery), autonomy (sense of personal control and ownership through, e.g., choice), and relatedness (sense of attachment to others through emotionally supportive relationships; Deci & Ryan, 2002). This study focused on competence, autonomy, and intrinsic motivation. Relatedness was included within the measure, but it was not included in subsequent analysis and discussion as previous research indicates this need is often supported through positive teacher-student relationships (Kiefer, Ellerbrock, & Alley, 2014). Thus, the use of a measure that looks strictly at instructional practices may not be effective in assessing student perceptions of the degree to which such practices support students’ sense of relatedness in the classroom.

Purpose of the Study

The purpose of the current study was to empirically create and validate a new measure to assess young adolescents’ perceptions of comprehension-focused LA classroom instructional practices on their needs and, through these, their intrinsic motivation to read in the middle grades LA classroom. This was accomplished by investigating the classroom instructional practices made available to teachers through adopted textbook series, and utilizing this information to select practices for inclusion in the measure. The measure, the Language Arts Reading Practices Survey (LARPS), was validated with a sample of middle grades students (N = 210) in a rural school district in the southeastern United States. This was part of a larger study that also included a parallel teacher form of the measure. Thus, students who participated in the study were in the LA classroom of one of eight participating teachers.

Defining Classroom Instructional Practices in LA

The extant research in literacy lacks consistency regarding what constitutes an instructional practice. Many studies do not explicitly define “classroom instructional practices” but give an implicit definition through examples (e.g., Pitcher et al., 2007; Wigfield et al., 2008). Other studies refer to specific activities in the LA classroom as instructional techniques (Hammerberg, 2004) or academic tasks (Matsumura, Correnti, & Wang, 2015). To provide clarity regarding the definition of classroom instructional practices, the current study built on the definitions of skills and strategies provided by Afflerbach, Pearson, and Paris (2008). They defined skills in reading, such decoding and comprehension, as automatic actions that result in successful meaning making from text, and they defined strategies as consciously chosen actions undertaken to support decoding and comprehension of texts (Afflerbach et al., 2008). Examples of strategies include making predictions and using context clues (Cunningham & Allington, 2011; Raphael, George, Weber, & Nies, 2014).

Using these definitions as a framework, we conceptualized classroom instructional practices in this study as specific activities and methods that enable teachers to instruct students in using strategies, assess how effectively students are applying the strategies, and assess students’ reading skills. The current study focused specifically on classroom instructional practices related to comprehension.

Method

Textbook Quantitative Content Analysis

Classroom instructional practices included in the initial item pool were selected through quantitative content analysis (Weber, 1990) of adopted middle grades LA textbooks series from California and Florida (Table 1). These states were chosen because they, together with Texas, make up about a third of the nation’s K–12 textbook market (Finn & Ravitch, 2004). Of the three states, California and Florida have the least overlap in textbook publishers on the adopted list of middle grades LA textbook series, while the adopted series for Texas includes overlap with both California and Florida. Thus, California and Florida provided for the most variety of publishers within the textbook analysis. In addition, the latest adoption for California was in 2009 (California Department of Education, 2015), which is comparable to Texas’s most recent adoption in 2010 (Texas Education Agency, 2015). Florida’s most recent adoption went into effect in the 2014–2015 academic year (Florida Department of Education, 2014). The study’s focus on textbooks also enabled an understanding of the instructional practices that these adopted instructional materials provide teachers.

Table 1. Textbook Series Used in Quantitative Content Analysis.

State	Citation
California	Glencoe Literature California Treasures, Teacher Edition. (2010). Columbus, OH: Glencoe/McGraw-Hill. (Course 1, 2, & 3)
	Holt Literature and Language Arts, Teacher’s Edition. (2010). Austin, TX: Holt, Rinehart, and Winston. (Introductory, first, and second course)
	McDougal Littell Literature, Teacher Edition. (2009). Evanston, IL: McDougal Littell. (Grade 6, 7, & 8)
Florida	Mirrors and Windows: Connecting with Literature, Annotated Teacher’s Edition. (2012). St. Paul, MN: EMC Publishing LLC. (Level I, II, & III)
	Florida Collections, Teacher’s Edition. (2015). Orlando, FL: Houghton Mifflin Harcourt Publishing Company. (Grade 6, 7, & 8)
	Pearson Common Core Literature Florida, Teacher’s Edition. (2015). Boston, MA: Pearson. (Grade 6, 7, & 8)

The analysis identified classroom instructional practices most frequently included within adopted series’ teachers’ editions. This quantitative content analysis required counting the occurrences of meaning units, including specific words, phrases, or content categories (Stemler, 2001; Weber, 1990). Due to the interactive and multifaceted nature of classroom instruction, some classroom instructional practices fit within more than one code (e.g., small group collaboration in completing a graphic organizer was coded both collaborative/small group activity and graphic organizer). There was minimal ambiguity in text coding, as classroom instructional practices coded within text were often identified directly by name (e.g., graphic organizer, small group discussion) or clearly identifiable by format (e.g., open-ended questions, multiple-choice questions). The PI conducted the full initial coding (Saldaña, 2013) through a manual page-by-page analysis. This first round of coding resulted in 37 initial codes.

Interrater reliability

An additional trained researcher coded a randomly selected 50-page section of one textbook using the codes created by the PI for purposes of interrater reliability. Cohen’s kappa was calculated to statistically determine the level of agreement beyond chance between the two coders for this section of text. There was substantial agreement per guidelines established by Landis and Koch (1977), κ = .798 (95% CI, .61 to .80), p = .000.

Focused coding

After initial coding, focused coding was conducted to organize initial codes into related categories and, when appropriate, combine multiple codes into a single code. For example, 3-column chart, character map, vocab-o-gram, and graphic organizers: other were combined into a single graphic organizer category. This resulted in 29 focused codes.

Once focused coding was completed, the PI calculated frequency counts for each code to determine trends in the textbooks. Individual item totals and page number totals per item were used to assure chosen practices were representative of practices presented within the textbooks analyzed and to address the concern that counting each occurrence equally might oversimplify interpretation of the data (Weber, 1990).

Once the textbook analysis was complete, the classroom instructional practices with the highest frequency counts (all more than 350 individual instances and located on at least 260 pages overall; Table 2) were included in the first draft of the measure. These cut-off points were selected based on a substantial drop in frequency after these points. For example, frequency by page total dropped from 260 to 198 occurrences, and full total frequency dropped from 353 to 282. There were few differences between these two methods of determining the most frequent classroom instructional practices (Table 2). Based on these cut-off points, 15 classroom instructional practices were included in the initial draft of the LARPS.

Table 2. Initial Item Pool for LARPS, Ranked by Overall Frequency.

Code	Name	Frequency rank by full total	Frequency rank by page total
9	Open-ended questions in textbook	1	2
10	Teacher-initiated questions	2	1
15	Graphic organizer	3	3
13	Whole class, teacher-mediated discussion	4	4
22	Multiple-choice questions	5	16
14	Collaborative discussions	6	5
6	Audio of text	7	6
16	Think-aloud modeling by teacher	8	7
20	Creative (non-essay) response to text	9	9
11	Take notes	10	8
18	Respond to text in essay format	11	10
8	Teacher read-aloud	12	11
2	Summarize	13	13
24	Small group/partner assignment	14	12
3	Student-generated questions	15	15

Expert panel feedback

The initial draft of the LARPS and a feedback form were sent to an expert panel consisting of two LA teachers with 35 combined years of experience, two LA researchers, one researcher specializing in SDT, and one measurement expert. The goal of the expert panel was to provide feedback regarding the classroom instructional practices identified and the language of the measure. Based on feedback from the expert panel, three classroom instructional practices were removed from the revised measure for the validation study (audio support, teacher read-aloud, summary). Summarizing was removed because the panel determined it to be a strategy and not an instructional practice (Kamil, 2004; Wharton-McDonald & Swiger, 2009). Teacher read-aloud and audio support were removed based on research identifying these practices as methods of supporting student fluency by providing models of expressive reading and proper pronunciation (Hurst & Griffity, 2015; Wolfson, 2008). Thus, their removal from the LARPS was appropriate because of the measure’s focus on comprehension practices. Expert panel feedback also resulted in changes to the language used in the descriptions of classroom instructional practices and statement stems, and to the formatting of the measure.

Cognitive interviews

Following revisions based on input from the expert panel, cognitive interviews (Willis, 1999) with middle grades students provided further evidence of both construct and response processes validity (American Educational Research Association, 2014). Because the cognitive interviews involved children, they were conducted after receiving IRB approval and parental consent for study participation, but prior to survey administration within the classrooms. Thus, no changes were made to the measure itself, but the protocol for survey administration included language addressing changes suggested by cognitive interviews to support participants’ understanding of the measure itself and how to respond to the measure. Four middle grades students (two high and two lower achieving readers) were nominated by teachers at one participating school. High achieving students were those who consistently earned A grades in their LA class, while lower achieving students consistently earned a C– or below in LA class. The four students each followed along in the LARPS as the PI read the measure aloud and answered student questions. The PI took notes regarding student questions. After students completed the measure, the PI used both scripted and spontaneous retrospective probes (Willis, 1999) to determine students’ understanding of items on the measure and their ability to complete the measure unaided. Changes suggested by the cognitive interviews included clarifying that collaborative discussions do not involve the creation of any product (more clearly differentiating it from partner/small group work), and clarifying the description of teacher think-aloud.

Validation Study

Participant demographics

Study participants were middle grades students not classified as struggling readers or English Language Learners at two schools in a rural school district in the southeastern United States. Among participants, the mean age was 12.37 years with ages ranging from 10 to 15. Fifty-two percent of participants were female. The ethnicity of the sample was 70.5% Caucasian, 15.2% other or mixed race, 11.4% African American, 2.4% Asian American, and 0.5% Native American. Slightly more than 26% of participants identified as Hispanic (Table 3).

Table 3. Demographic Characteristics of Participants (N = 210).

	School A		School B		Total sample
	n	%	n	%	N	%
Gender
Male	89	47.6	11	47.8	100	47.6
Female	98	52.4	12	52.2	110	52.4
Grade
6	96	51.3	15	65.2	111	52.9
7	36	19.3	0	0.0	36	17.1
8	55	29.4	8	34.8	63	30.0
Age
10	1	0.5	0	0.00	1	0.5
11	28	15.0	3	13.0	31	14.8
12	70	37.4	11	47.8	81	38.6
13	45	24.1	2	8.7	47	22.4
14	40	21.4	7	30.4	47	22.4
15	3	1.6	0	0.0	3	1.4
Race
Black/African American	21	11.2	3	13.0	24	11.4
Asian/Pacific Islander	4	2.1	1	4.3	5	2.4
White	131	70.1	17	73.9	148	70.5
Native American	1	0.5	0	0.0	1	0.5
Other/Mixed Race	30	16.0	2	8.7	32	15.2
Ethnicity
Hispanic	48	25.7	7	30.4	55	26.2
Other	139	74.3	16	69.6	155	73.8

Measures

Student participants completed a demographic form followed by two measures: the LARPS and the MRQ (Wigfield & Guthrie, 1997).

Demographic form

Student demographic information was collected via a demographic form that was part of the survey packet.

LARPS—student form

The LARPS is a 48-item measure of student perceptions of LA classroom instructional practices. The measure consists of a separate subscale for each of the needs (competence and autonomy) and intrinsic motivation, as well as a subscale for the frequency with which classroom instructional practices are used in the classroom. The measure uses a 6-point scale (1 = not at all; 5 = very much; 6 = I don’t know) to indicate to what degree students perceive specific practices within their LA classroom as supporting their needs and intrinsic motivation to read. For each practice, students indicated how often the practice was used and the degree to which that practice made them feel they could be successful readers (competence) and allowed them options for sharing their reading (autonomy). In addition, students indicated whether that practice supported their intrinsic motivation to read. For purposes of data analysis, I don’t know responses were coded as missing data (see Appendix for measure information).

Motivations for reading questionnaire (MRQ)

The MRQ (Wigfield & Guthrie, 1997) is a 53-item measure of students’ motivation to read and contains 11 subscales with two to seven items each. The subscales are reading efficacy, challenge, curiosity, involvement, importance of reading, work avoidance, competition, recognition for reading, reading for grades, social reasons, and compliance. Student responses are on a 4-point Likert scale (1 = very different from me to 4 = a lot like me; Wigfield & Guthrie, 1997). Although this measure was originally created for use with fourth- and fifth-grade students, it has been successfully used with middle grades populations (Paige, 2011; Unrau & Schlackman, 2006). Analyses of the measure have indicated reliabilities for the various subscales ranging from .59 (work avoidance) to .81 (recognition; Mucherah & Ambrose-Stahl, 2014; Mucherah & Yoder, 2008). The results of this self-report survey were used to support concurrent validity for the LARPS, as both measures address elements of student motivation to read through different theoretical frameworks.

Data screening

Visual scans of completed survey packets to check for completeness assured minimal missing data on the measures except for I don’t know responses on the LARPS, which were treated as missing data. Use of mean scores in place of missing data did not result in significant differences in results. Thus, missing data were retained as such, and pairwise deletion was used to insure all available data were included in analyses. The data were screened for outliers using SAS version 3.4. No outliers were identified for the LARPS (Table 4). Normality for the LARPS was examined, and all skewness and kurtosis statistics were less than ±3, indicating a trend of normal distribution (Kline, 2010).

Table 4. Descriptive Statistics for the LARPS.

Item	N	Mean	SD	Skew	Kurtosis
(1) Open-ended textbook questions freq	210	3.80	0.87	−0.778	0.910
Competence	204	3.62	1.05	−0.225	−0.507
Autonomy	201	3.40	1.17	−0.308	−0.697
Intrinsic motivation	196	3.07	1.36	−0.028	−1.081
(2) Open-ended questions asked by teacher freq	209	4.13	1.05	−1.433	1.815
Competence	197	3.66	1.17	−0.591	−0.358
Autonomy	200	3.52	1.19	−0.428	−0.482
Intrinsic motivation	197	3.19	1.36	−0.161	−1.081
(3) Multiple-choice questions freq	208	4.05	0.67	−0.729	2.003
Competence	204	3.73	1.09	−0.583	−0.179
Autonomy	200	3.40	1.24	−0.353	−0.750
Intrinsic motivation	203	3.02	1.46	−0.084	−1.309
(4) Teacher-directed, whole-class discussion freq	210	4.23	0.93	−1.494	2.416
Competence	205	3.76	1.16	−0.595	−0.444
Autonomy	198	3.73	1.18	−0.660	−0.350
Intrinsic motivation	200	3.22	1.38	−0.252	−1.074
(5) Collaborative discussion freq	209	3.69	0.94	−0.897	1.119
Competence	205	3.71	1.17	−0.616	−0.306
Autonomy	202	3.73	1.21	−0.610	−0.574
Intrinsic motivation	202	3.36	1.39	−0.278	−1.123
(6) Partner/small group work freq	210	3.56	1.12	−0.723	0.401
Competence	206	3.71	1.20	−0.743	−0.160
Autonomy	203	3.69	1.24	−0.711	−0.417
Intrinsic motivation	200	3.41	1.42	−0.415	−1.075
(7) Graphic organizer freq	210	3.23	0.92	−0.444	−0.141
Competence	197	3.47	1.21	−0.340	−0.674
Autonomy	197	3.38	1.22	−0.325	−0.717
Intrinsic motivation	198	3.06	1.37	−0.122	−1.127
(8) Take notes freq	208	3.35	1.23	−0.419	−0.628
Competence	194	3.66	1.28	−0.631	−0.613
Autonomy	192	3.41	1.26	−0.315	−0.860
Intrinsic motivation	185	3.16	1.44	−0.179	−1.230
(9) Student-generated questions freq	209	2.22	1.23	0.672	−0.579
Competence	171	3.16	1.36	−0.144	−1.061
Autonomy	172	3.16	1.39	−0.203	−1.185
Intrinsic motivation	175	2.86	1.43	0.099	−1.211
(10) Teacher think-aloud freq	210	3.60	1.24	−0.643	−0.504
Competence	195	3.63	1.24	−0.559	−0.582
Autonomy	195	3.37	1.28	−0.313	−0.908
Intrinsic motivation	193	3.31	1.41	−0.352	−1.083
(11) Creative response to text freq	208	3.13	1.15	−0.432	−0.567
Competence	197	3.65	1.16	−0.599	−0.221
Autonomy	193	3.50	1.17	−0.424	−0.532
Relatedness	192	3.35	1.29	−0.346	−0.804
Intrinsic motivation	193	3.32	1.40	−0.382	−1.086
(12) Essay response to text freq	208	3.36	0.87	−0.012	−0.342
Competence	202	3.55	1.23	−0.521	−0.526
Autonomy	202	3.40	1.26	−0.375	−0.828
Intrinsic motivation	200	3.02	1.46	−0.074	−1.286

Note. freq indicates statistics for frequency students perceived each practice as being used.

Findings

Measure Validity

Correlations between subscale items of the LARPS and subscales of the MRQ (Wigfield & Guthrie, 1997) associated with intrinsic motivation were calculated to determine support for the validity of the LARPS as a measure of student motivation to read. Before determining correlations between items on the LARPS subscales and subscales of the MRQ (Wigfield & Guthrie, 1997), Cronbach’s alpha was calculated for all subscales of the MRQ. All subscales of the MRQ related to intrinsic motivation indicated acceptable reliability, ranging from .683 (social reasons for reading) to .830 (challenge).

LARPS subscale item correlations with the MRQ

Correlations between items on the various subscales of the LARPS and subscales of the MRQ support the concurrent validity of the LARPS as a measure of adolescents’ intrinsic motivation to read. Specifically, items from the LARPS tended to be significantly positively related to subscales of the MRQ which are associated with intrinsic motivation to read, including curiosity, involvement in reading, challenge, and importance of reading (Guthrie, Wigfield, Metsala, & Cox, 1999; Varuzza et al., 2014).

Competence subscale

Items from the competence subscale of the LARPS were all significantly positively correlated with the involvement subscale of the MRQ (Table 5). Correlations for involvement ranged from .147 (graphic organizers) to .298 (small group work). All LARPS items except for creative writing response to text were significantly positively associated with the importance of reading subscale. Students who reported higher beliefs in the importance of reading were more likely to endorse all items except creative writing as supportive of their sense of competence in reading. Graphic organizers and essay response to text were the only LARPS items not associated with the challenge subscale of the MRQ. In addition, all LARPS competence items were positively associated with the self-efficacy subscale of the MRQ. Thus, students who reported high levels of self-efficacy were more likely to endorse classroom instructional practices on the LARPS as supporting their need for competence. This finding supports the concurrent validity of this specific subscale of the LARPS, as reading self-efficacy has been defined as confidence in one’s reading ability (Wigfield et al., 2008).

Table 5. Correlations between LARPS Competence Subscale Items and MRQ.

	Self-efficacy	Challenge	Curiosity	Importance	Involvement	Competition	Recognition	Social reasons
Open-ended text questions	.410**	.203**	.090	.235**	.206**	.157*	.271**	.250**
Open-ended teacher questions	.358**	.142*	.056	.216**	.169*	.084	.183*	.206**
Multiple-choice questions	.362**	.214**	.151*	.265**	.230**	.144*	.267**	.290**
Teacher-directed discussion	.308**	.253**	.173*	.186**	.261**	.180*	.259**	.295**
Collaborative discussion	.299**	.188**	.221*	.250**	.182**	.207**	.301**	.233**
Small group work	.345**	.303**	.267*	.310**	.298**	.230**	.334**	.269**
Graphic organizer	.328**	.121	.113	.222**	.147*	.168*	.226**	.225**
Take notes	.278**	.185**	.102	.190**	.183*	.142*	.280**	.313**
Student-generated questions	.347**	.258**	.153*	.245**	.270**	.216**	.284**	.272**
Teacher think-aloud	.352**	.279**	.184*	.292**	.294**	.142*	.328**	.263**
Creative writing	.323**	.203**	.078	.139	.277**	.179*	.267**	.261**
Essay	.223**	.134	.077	.179*	.168*	.204**	.253**	.247**
M (SD)	2.86 (.73)	2.60 (.81)	2.82 (.68)	2.78 (.90)	2.77 (.77)	2.71 (.71)	2.51 (.75)	1.77 (.60)

Note. See Table 4 for LARPS competence item means and N. N range 171–208. **p < .01 (2-tailed); *p < .05 (2-tailed)

Autonomy subscale

The autonomy subscale of the LARPS had fewer significant associations with subscales of the MRQ (Table 6). All but one item (student-generated questions) were positively associated with the self-efficacy subscale of the MRQ. Thus, students who reported high levels of self-efficacy were more likely to endorse classroom instructional practices on the LARPS as supportive of their autonomy. This finding was similar to the results of previous research that found positive associations between student perceptions of teacher autonomy support and their academic competence (i.e., self-efficacy; Guay, Ratelle, Larose, Vallerand, & Vitaro, 2013). Fewer items were significantly associated with subscales of the MRQ related to intrinsic motivation; 8 of 12 items were positively associated with the involvement in reading subscale, while only 4 of 12 were positively associated with the challenge subscale. This lack of significant association may be due to the absence of subscales on the MRQ that directly address autonomy or autonomous motivation for reading.

Table 6. Correlations between LARPS Autonomy Subscale Items and MRQ.

	Self-efficacy	Challenge	Curiosity	Importance	Involvement	Competition	Recognition	Social reasons
Open-ended text questions	.202**	.081	.053	.095	.100	.161*	.169*	.137
Open-ended teacher questions	.223**	.074	.114	.160*	.138	.124	.182**	.125
Multiple-choice questions	.189**	.036	.065	.122	.148*	.132	.213**	.202**
Teacher-directed discussion	.205**	.118	.101	.079	.206**	.147*	.178*	.190**
Collaborative discussion	.233**	.120	.109	.132	.164*	.127	.202**	.157*
Small group work	.166**	.173*	.126	.134	.205**	.180*	.208**	.071
Graphic organizer	.237**	.143*	.117	.174*	.204**	.141*	.200**	.181*
Take notes	.198**	.152*	.055	.097	.173*	.070	.204**	.204**
Student-generated questions	.128	.014	−.012	.064	.071	.130	.210**	.155*
Teacher think-aloud	.261**	.163*	.094	.182*	.165*	.039	.197**	.185*
Creative writing	.229**	.080	.050	.075	.136	.127	.145*	.139
Essay	.184**	.123	.068	.152*	.162*	.221**	.231**	.252**
M (SD)	2.86 (.73)	2.60 (.81)	2.82 (.68)	2.78 (.90)	2.77 (.77)	2.71 (.71)	2.51 (.75)	1.77 (.60)

Note. See Table 4 for LARPS autonomy item means and N. N range 172–203. **p < .01 (2-tailed); *p < .05 (2-tailed)

Intrinsic motivation subscale

All intrinsic motivation items significantly correlated with the self-efficacy subscale of the MRQ and with subscales related to intrinsic motivation (challenge and involvement; Table 7). Associations between self-efficacy and intrinsic motivation aligned with previous research that found similar positive associations for both struggling and advanced readers in the middle grades (Klauda & Guthrie, 2015). All but one item (student-generated questions) correlated with the importance of reading subscale of the MRQ, which is also associated with intrinsic motivation (Guthrie et al., 1999). Therefore, students who highly endorsed a belief in the importance of reading were more likely to report all items except student-generated questions as supportive of their intrinsic motivation to read. These associations between subscales of the MRQ related to intrinsic motivation support the concurrent validity of the LARPS as a measure of early adolescents’ intrinsic motivation to read.

Table 7. Correlations between LARPS Intrinsic Motivation Subscale Items and MRQ.

	Self-efficacy	Challenge	Curiosity	Importance	Involvement	Competition	Recognition	Social reasons
Open-ended text questions	.500**	.378**	.325**	.365**	.363**	.239**	.384**	.344**
Open-ended teacher questions	.419**	.365**	.307**	.258**	.361**	.119	.312**	.252**
Multiple-choice questions	.398**	.244**	.203**	.231**	.317**	.137	.241**	.210**
Teacher-directed discussion	.301**	.298**	.217**	.186**	.318**	.068	.228**	.194**
Collaborative discussion	.350**	.289**	.230**	.258**	.298**	.146*	.261**	.221**
Small group work	.325**	.259**	.184**	.285**	.296**	.206**	.276**	.230**
Graphic organizer	.301**	.237**	.199**	.247**	.232**	.132	.250**	.228**
Take notes	.243**	.177*	.114	.165*	.212**	.080	.275**	.211**
Student-generated questions	.275**	.179*	.102	.143	.211**	.029	.187*	.188*
Teacher think-aloud	.239**	.172*	.045	.203**	.192**	.119	.214**	.138
Creative writing	.258**	.253**	.161*	.227**	.306**	.113	.266**	.304**
Essay	.216**	.205**	.108	.227**	.188**	.501**	.256**	.267**
M (SD)	2.86 (.73)	2.60 (.81)	2.82 (.68)	2.78 (.90)	2.77 (.77)	2.71 (.71)	2.51 (.75)	1.77 (.60)

Note. See Table 4 for LARPS intrinsic motivation item means and N. N range 175–203. **p < .01 (2-tailed); *p < .05 (2-tailed)

Reliability of the LARPS

Cronbach’s alpha was calculated using SAS version 3.4 for each subscale of the LARPS. The Cronbach’s alpha for all subscales indicated a high level of reliability (Competence = .940; Autonomy = .923; Intrinsic motivation = .955).

Item-total correlations for each subscale further support the reliability of the LARPS. Items on the competence subscale had a moderate to strong item-total correlation (ranging from .625 to .790). This pattern was also evident in analysis of the autonomy subscale, with item-total correlations ranging from .593 to .755. Finally, the intrinsic motivation subscale (item total correlations ranging from .683 to .827) indicated support for the reliability of the LARPS. Analyses of all subscales did not indicate the need for removal of any items.

Discussion

The initial draft of the LARPS was modified based on iterative feedback from panels of experts, resulting in a measure consisting of 12 classroom instructional practices. The measure was implemented with a sample of middle grades LA students to establish validity and reliability of the modified measure. The results of this validation implementation provided preliminary support for the reliability and validity of the LARPS.

Associations between the LARPS and Subscales of the MRQ

Student-generated questions, an item on the LARPS, had few associations with subscales of the MRQ related to intrinsic motivation. Interestingly, students reported this classroom instructional practice as used least frequently, averaging approximately once per semester (mean of 2.22 compared to next lowest mean of 3.13 for creative writing response). Perhaps due to its perceived infrequent use, 42 students selected I don’t know for at least one item indicating their perceptions of the degree to which this practice met their needs (the highest use of this response on the measure). This may have contributed to a lack of association among student perceptions of this practice and subscales of the MRQ such as challenge, curiosity, and importance of reading. The infrequency with which student-generated questions was reported was not surprising because teacher support and scaffolding is required for students to engage effectively in this practice (Humphries & Ness, 2015; McNamara, Ozuru, Best, & O’Reilly, 2007). As one participating teacher commented to the PI, she would like to provide more opportunities for her students to generate their own questions, but “there isn’t time.” This lack of time (as perceived by teachers) is seen in previous research regarding LA teachers’ perceptions of motivating classroom instruction (Delaney et al., 2014).

There were positive associations among student perceptions of collaborative discussion and small group work as supportive of their intrinsic motivation and MRQ subscales related to intrinsic motivation. Previous research (Guthrie & Klauda, 2014) indicated collaboration is positively associated with students’ intrinsic motivation. These findings relate to support for collaboration as a general practice used within an intervention consisting of multiple elements, making it difficult to isolate the associations between intrinsic motivation and specific classroom practices that support collaboration. The results of the current study show this association in regards to specific classroom instructional practices (small group work and collaborative discussion) as opposed to an intervention program consisting of multiple instructional practices.

Across all subscales of the LARPS, all items were positively associated with the self-efficacy subscale of the MRQ, except for student-generated questions on the autonomy subscale. As self-efficacy in reading is conceptualized as one’s confidence in comprehension and language skills (Wigfield, Guthrie, Tonks, & Perencevich, 2004), this positive association supports the validity of the competence subscale of the LARPS. As SDT considers the satisfaction of all three needs (competence, autonomy, and relatedness) as necessary for optimal outcomes, including higher levels of intrinsic motivation (Deci & Ryan, 2001), these positive associations across multiple subscales make theoretical sense.

Additional research investigating these associations could improve understanding of the ways in which recognition in the LA classroom is associated with and influences students’ perceptions of need fulfillment within the classroom, providing teachers with guidance in utilizing recognition to support their students’ needs and self-determined motivation more effectively.

Contributions to the Literature

Research has indicated associations between adolescents’ motivation to read and their reading achievement (Guthrie et al., 2013; Kelley & Decker, 2009), as well as declines in adolescents’ motivation to read during the middle grades (Kelley & Decker, 2009; Lepper et al., 2005). Little research has investigated associations among adolescents’ motivation to read and classroom instructional practices, despite the importance of these practices within this context. The current study sought to gain an understanding of the degree to which classroom instructional practices are perceived as supporting early adolescents’ intrinsic motivation to read as well as their needs for competence and autonomy through the development and validation of the LARPS.

Theoretical implications

Previous research, as well as SDT, highlight the importance of meeting early adolescents’ needs for competence, autonomy, and relatedness to support their motivation within the classroom (Deci & Ryan, 2002). As intrinsic motivation is associated with reading achievement (Becker, McElvany, & Kortenbruck, 2010; Froiland & Oros, 2013), knowledge of these associations between classroom practices and intrinsic motivation is needed to identify specific practices that support students’ needs and intrinsic motivation. Thus, the LARPS has theoretical implications for understanding how classroom instructional practices meet students’ needs for competence and autonomy, and support their intrinsic motivation to read in the middle grades LA classroom. Additional study with larger, more diverse samples that considers possible moderating variables at the teacher and school levels is needed to confirm the reliability and concurrent validity of the LARPS.

Student perceptions of the learning environment and the degree to which their needs for competence and autonomy are met within that environment may vary by gender (Koth, Bradshaw, & Leaf, 2008; Wentzel, Battle, Russell, & Looney, 2010), grade level (Katz et al., 2010; Wentzel et al., 2010), race (Koth et al., 2008), and identified disability (Ferguson, Hanreddy, & Draxton, 2011). Additional factors, such as students’ motivations for reading (e.g., social reasons for reading, need for recognition, and value of reading) and reading self-efficacy may also influence student perceptions of the learning environment and the degree to which aspects of that environment meet their needs and support their intrinsic motivation to read (DeNaeghel et al., 2012). Through analysis of the relationships between subscales of the LARPS and the MRQ, the current study provides preliminary findings indicating classroom instructional practices may be associated with student needs and intrinsic motivation in unique ways based on the individual student’s self-efficacy and motivations for reading. For example, students who reported high self-efficacy were more likely to endorse open-ended textbook questions as supportive of their intrinsic motivation to read (r = .500, p < .01). Although students who perceive themselves as capable readers may view this practice as motivating, this may not be true for students who view themselves as less efficacious in reading. Thus, it may be important to consider early adolescents’ baseline characteristics, including current levels of self-efficacy, as well as specific motivations for reading, such as value of reading, to select classroom instructional practices that will best be supportive of their needs and intrinsic motivation to read. The results of this study indicate specific classroom practices may interact with student baseline characteristics to influence the degree to which their needs and intrinsic motivation are supported within the classroom. This is supported by SDT, which posits that contextual elements both within the individual, such as baseline self-efficacy, and from outside sources, such as classroom instructional practices, result in differences in the degree to which they perceive their needs being met and are, therefore, intrinsically motivated (Ryan & Deci, 2000). Associations among student motivations for reading and perceptions of the degree to which LA classroom instructional practices support students’ needs and self-determined motivation may provide a more nuanced and context-specific understanding of the interplay between these variables.

Practical implications

Knowledge that can guide teachers in effectively supporting students’ intrinsic motivation through their needs for competence and autonomy, such as that provided by student responses to the LARPS, may assist in reversing the trend of decreasing motivation among students in the middle grades. Through student responses to the LARPS, teachers can identify classroom instructional practices perceived by students as supportive of their needs and intrinsic motivation. For practitioners, examining the extent to which classroom instructional practices are responsive to and meet students’ needs may provide insight into what literacy classroom instructional practices promote students’ intrinsic motivation to read. This insight is important, as higher levels of self-determined motivation to read are associated with adaptive outcomes in reading (Mucherah & Yoder, 2008; Schaffner, Schiefele, & Ulferts, 2013). For example, mean scores on the LARPS indicate students viewed instructional practices such as collaborative discussion, teacher directed whole-class discussion, and teacher think-aloud as somewhat supportive of their needs for competence, autonomy, relatedness, and their intrinsic motivation. Associations discussed previously, such as those between self-efficacy and student perceptions of classroom practices as supportive of intrinsic motivation, point to other student variables that can guide teacher selection of instructional practices. Teachers can use such insight to select practices more purposefully within the middle grades LA classroom context.

Limitations

Despite several theoretical and practical contributions and strengths of the current study, there are several limitations that must be considered. Three primary limitations discussed below include the use of self-reported data, convenience sampling, and additional issues of population validity.

The use of self-reported student data was necessary for measuring perceptions of instructional practices within the LA classroom. Using self-reported data has several limitations. First, social desirability may influence participant responses (Baker & Wigfield, 1999; Traugott, 2004). For example, research has indicated that friends’ attitudes toward reading may influence young adolescents (Merga, 2014), and this influence may cause participants to respond to measures regarding reading in a manner that reflects the attitude of their friends more than their own views. To minimize social desirability, confidentiality of responses was emphasized prior to and throughout survey administration. A second limitation of self-report measures is item interpretation; participants may not understand abstract vocabulary used within the measure (Fulmer & Frijters, 2009). To minimize this concern, the PI used expert panel review by middle grades LA teachers and cognitive interviews with middle grades students to identify and address possible points of confusion. A third limitation is the use of a Likert format, which has been criticized for providing conceptually inaccurate scoring and a tendency for responses to be biased toward the positive end of the scale (Fulmer & Frijters, 2009). The use of a Likert format response may lead to an over-generalized view of the phenomena under investigation and ignore other variables that may be mediators and/or moderators, which may confound the results (Fulmer & Frijters, 2009). Future research that uses multiple data sources, such as participant interviews and classroom observations, may allow for triangulation of data and reduce the limitations of utilizing self-reports.

The use of a convenience sample of students is another limitation. Additional contextual factors, such as each school’s level of emphasis on literacy and school grade, were not taken into consideration when selecting participating schools for this study. School-wide cultural factors, such as the emphasis placed on reading and literacy as valued activities, may influence the motivations and attitudes of students and teachers and, thus, should be considered in future research.

An additional limitation is the small sample of students in this study. Future research utilizing the LARPS with a larger, more diverse sample will allow for more precise analysis of the reliability of the LARPS and may provide further support for the use of the measure to reliably assess student perceptions of the LA classroom.

Additional limitations include the unknown stability of the LARPS and the use of student data from a single time point, providing a snapshot of student perceptions within the LA classroom. Future research investigating changes in students’ perceptions over time, as well as differences in perception by grade level, may reveal additional insights. This is important for several reasons. First, there are differences in students’ needs across developmental periods (Eccles, 1999). For example, young adolescents often indicate an increased need for autonomy (Eccles et al., 1999; Eccles & Roeser, 2009). Second, academic tasks and expectations differ across grade levels, a fact that is evident in educational standards. Thus, some instructional practices used in a classroom for young adolescents may differ from those used with younger or older students.

Appendix. LARPS—Sample Formatted Item and Additional Item Verbiage

(1) Respond to open-ended textbook questions: Students respond, either orally or in writing, to open-ended questions (short answer, not multiple choice) found in the textbook.

Table

		Daily	Weekly	1–2 times a month	1–2 times a semester	Never
This is used in my LA classroom:		⃝	⃝	⃝	⃝	⃝
	A lot 5	4	Somewhat 3	2	Not at all 1	I don’t know
When I do this in my LA class:
I feel confident in my ability to be a successful reader.	⃝	⃝	⃝	⃝	⃝	⃝
I feel like I have options for sharing what I have read.	⃝	⃝	⃝	⃝	⃝	⃝
I participate because it makes reading enjoyable.	⃝	⃝	⃝	⃝	⃝	⃝

• (2) Respond to open-ended questions asked by teacher: Students respond to open-ended questions (short answer, not multiple choice), either orally or in writing, asked by the teacher (not from the textbook).

• (3) Respond to multiple-choice questions: Students respond to multiple-choice questions about the textbook. Such questions often have approximately four possible answers listed.

• (4) Class discussion: Students and teacher discuss the texts as a large group. This may include discussion of questions from the text or questions asked by the teacher.

• (5) Collaborative discussion: Students collaboratively discuss questions or ideas related to the text with peers in pairs or in small groups.

• (6) Partner/small group work: Students work with one or more other students to create a product in response to a text. The product may include a written response to questions about the text, a graphic organizer, a skit or play, poster, or any other product that can be shared with others once completed.

• (7) Graphic organizer: Students are given or asked to create a picture or chart to organize specific information about the text. This may be completed individually or in a small group.

• (8) Note taking: Students take notes about the text read. Notes may be made in a reader’s notebook, foldable, or through annotations within the text. Unlike a graphic organizer, these notes are not organized by boxes or other graphics.

• (9) Student-generated questions: Students create their own questions about the text. These questions may be answered through discussion with peers or in writing by the student.

• (10) Teacher think-aloud: The teacher reads aloud and models what good readers do, such as asking questions about the text, making inferences, summarizing, etc.

• (11) Creative response to text: Students respond to the text through creative writing, such as writing a journal entry or narrative from a character’s point of view, writing a narrative that takes place in the story setting, or other format.

• (12) Essay response to text: Students respond to the text by writing an informational, persuasive, or narrative essay (multiple paragraphs) that uses information from the text to support ideas presented in the essay.