Examining the Psychometric Properties of the Homework Management Scale for High School Students in China

Abstract

This study reports on the psychometric evaluation of the Chinese version of the Homework Management Scale (HMS). The HMS was designed to assess students’ homework management strategies. Based on a randomized split of 884 high school students in China, we conducted exploratory factor analysis on Group 1 (n = 442) and confirmatory factor analysis on Group 2 (n = 442). The factor structure of the Chinese version of the HMS was consistent with the original one developed in the United States (i.e., the existence of five separate, yet related, subscales). Reliability coefficients of the HMS and its subscales were in the adequate to good range. Finally, the HMS and its subscales, as predicted, were related to relevant homework measures in the theoretically expected directions. These findings suggest that the Chinese version of the HMS is a valid multidimensional measure for homework management. Suggestions for its application and future research are provided.

Keywords

homework homework management scale scale adaption validity self-regulation

Introduction

Homework plays a special role in children’s life. It removes learning from the classroom, where a teacher closely arranges the learning environment and monitors children’s activities. It brings learning to children’s daily life, where learning coexists with other leisure and maintenance activities, where they have more personal discretion concerning whether, when, and how to complete the homework tasks. It is not surprising that self-regulatory aspects of homework have often been assumed (e.g., better time organization and greater self-direction; Corno, 2000; Corno & Xu, 2004; Xu, 2004; Zimmerman & Kitsantas, 2005). It is surprising to note, however, that research on homework focuses on its academic outcomes (Cooper, Robinson, & Patall, 2006; Zimmerman & Kitsantas, 2005). One explanation for inadequate attention to homework’s nonacademic outcomes is the lack of valid instruments to measure homework behavior (Cooper et al., 2006; Power, Dombrowski, Watkins, Mautone, & Eagle, 2007).

Among major self-regulation models in education (e.g., the cognitive and sociocultural aspects of self-regulation), Corno’s model on volitional control (Boekaerts & Corno, 2005; Corno, 2004) is particularly pertinent to homework, as this model is mainly concerned with the implementation of intentions that occur after goals are set. Specifically, it is characterized by the self-regulation activities of purposive striving, including, for example, prioritizing, bypassing obstacles, managing resources, budgeting time, and regulating emotional states (Corno, 2004).

Doing homework often means complying with the demand of completing externally imposed academic tasks (Corno, 2000), in the sense that teachers typically set homework goals, and the main charge for students is to implement and follow through homework intention. In doing so, students are required to navigate the demands of doing homework, which often require volitional control. They are required to manage homework, including, for example, planning, organizing the workspace, sustaining the strength of homework intention, handling distractions, and debilitating unwanted emotions surrounding homework assignments (Xu, 2008b).

The initial psychometric evidence for the Homework Management Scale (HMS) was conducted with 987 U.S. 11th graders (Xu, 2008b). The results revealed that the HMS comprised five separate, yet related, factors: Arranging the Environment, Managing Time, Handling Distraction, Monitoring Motivation, and Controlling Emotion. Specifically, exploratory factor analyses (EFA) yielded a five-factor solution that accounted for 60.63% of the variance in the HMS scores, in line with the hypothesized five features of homework management. After including two error covariances—one between two items relating to monitoring motivation and another between two items relating to arranging the environment—confirmatory factor analysis (CFA) indicated an adequate fit of the final model to the data for rural high school students (*CFI [robust comparative fit index] = .942; SRMR [standardized root mean square residual] = .056; *RMSEA [robust root mean square error of approximation[ = .049; 90% CI [confidence interval] = [.040, .057]) and urban high school students (*CFI = .951; SRMR = .055; *RMSEA = .041; 90% CI = [.029, .051]).

Taken together, the HMS appears to be a factorially valid measure of high school students’ effort to independently managing homework in the United States. It may hold promise as a measure of homework management for students in other countries in that homework is a common instructional practice extending across national boundaries (Cooper et al., 2006; Warton, 2001). In addition, the demands associated with doing homework are similar across different countries (e.g., managing time and handling distraction; Kouzma & Kennedy, 2002; Verma, Sharma, & Larson, 2002; Xu & Corno, 1998). Finally, the items in the HMS are not country specific.

Thus, it would be important to adapt the HMS for high school students in different countries and to examine its psychometric properties. One promising starting point is to conduct a validation study of the HMS in China. Chinese culture places a high value on education and academic achievement (Li & Kirkup, 2007). In addition, Chinese beliefs in human malleability and self-improvement result in a strong emphasis on effort in educational activities (Chen & Uttal, 1988). Consequently, compared with U.S. students, Chinese students often hold more positive attitudes toward homework (Chen & Stevenson, 1989; Hong, Wan, & Peng, 2011). As students’ attitudes toward homework may influence how they approach homework (e.g., managing homework independently), it would be interesting to examine the applicability of the HMS to high school students in China.

In addition, the previous study by Xu (2008b) used a general indicator of homework management (i.e., homework management across different subject areas). As some recent studies have tapped into domain-specific aspects of homework (e.g., effort and self-regulation; Hong, Peng, & Rowell, 2009; Trautwein, Ludtke, Schnyder, & Niggli, 2006) and as Chinese students outperform their U.S. peers in math and excel in many international assessments of math achievement (Wang, 2004), it would be informative to focus on homework management in one important domain (i.e., math) in the present investigation.

The aim of the present study was to adapt the HMS for high school students in China on the domain of math homework and to examine its psychometric properties. Specifically, the purposes of the present study were to examine (a) the factor structure of the HMS with high school students in China, (b) internal consistency reliability evidence for the HMS, and (c) construct and concurrent validity by testing the relationship between scores on the HMS and scores assessing relevant homework measures (e.g., homework purpose and behavior).

Method

Participants

The participants were 884 eleventh-grade public school students in southeast China. Specifically, of these participants, 472 were females (53.4%) and 412 were males (46.6%). The mean educational level for their parents was 15.0 years (SD = 2.5).

With respect to math homework, almost all participants (98.4%) reported that they received math homework assignments five or more days a week. In addition, they reported that they spent 71 min doing math homework on a typical day (SD = 34).

Instrument

The HMS consists of 22 items using a 5-point response format, in which students are asked to choose a response from 1 (never), 2 (rarely), 3 (sometimes), 4 (often), or 5 (routinely). It comprises five subscales: (a) Arranging the Environment (5-item subscale, for example, “find a quiet area”), (b) Managing Time (4-item subscale, for example, “set priority and plan ahead”), (c) Handling Distraction (5-item subscale, for example, “stop math homework to send or receive ‘instant messaging’”), (d) Monitoring Motivation (4-item subscale, for example, “find ways to make math homework more interesting”), and (e) Controlling Emotion (4-item subscale, for example, “tell myself to calm down”). All five items in one subscale (i.e., Handling Distraction) were reverse scored (see Table 1).

Table 1.

Rotated Factor Pattern (Structure) Matrix for the HMS (Group 1, N=442).

Item		Environment	Time	Distraction	Motivation	Emotion
1	Locate the materials I need for my homework	.477 (.502)	.078 (.211)	−.016 (.073)	.084 (.160)	−.020 (.096)
2	Find a quiet area	.668 (.716)	.075 (.259)	.119 (.245)	.064 (.180)	.002 (.168)
3	Remove things from the table	.869 (.822)	−.109 (.088)	−.107 (.049)	−.060 (.037)	.042 (.096)
4	Make enough space for me to work	.859 (.832)	−.066 (.133)	−.059 (.098)	−.060 (.053)	.050 (.123)
5	Turn off the TV	.398 (.461)	.120 (.224)	.183 (.269)	−.044 (.071)	.052 (.162)
6	Set priority and plan ahead	.146 (.327)	.665 (.740)	.011 (.076)	.147 (.347)	−.001 (.227)
7	Keep track of what remains to be done	.174 (.346)	.763 (.799)	.029 (.017)	.075 (.261)	−.109 (.107)
8	Remind myself of the available remaining time	−.131 (.172)	.822 (.822)	.005 (.049)	−.014 (.223)	.045 (.219)
9	Tell myself to work more quickly when I lag behind	−.144 (.049)	.685 (.696)	−.037 (.033)	−.050 (.222)	.279 (.381)
10	Find ways to make math homework more interesting	.046 (.162)	.115 (.296)	.052 (.081)	.690 (.697)	−.085 (.227)
11	Praise myself for good effort	−.013 (.095)	−.041 (.210)	−.056 (−.005)	.921 (.920)	.037 (.370)
12	Praise myself for good work	−.051 (.053)	−.069 (.175)	−.051 (−.002)	.905 (.901)	.064 (.380)
13	Reassure myself that I am able to do math homework when I feel it is too hard	−.021 (.130)	.075 (.284)	.099 (.201)	.507 (.654)	.323 (.556)
14	Tell myself not to be bothered with previous mistakes	.035 (.177)	.113 (.177)	−.005 (.185)	.045 (.372)	.752 (.798)
15	Tell myself to pay attention to what needs to be done	.005 (.149)	.127 (.149)	−.026 (.168)	.016 (.358)	.795 (.824)
16	Tell myself to calm down	.085 (.212)	.020 (.212)	.089 (.285)	−.047 (.275)	.778 (.798)
17	Cheer myself up by telling myself that I can do it	.036 (.147)	−.088 (.137)	−.011 (.176)	.193 (.461)	.744 (.802)
18	Daydream during a math homework session^a	−.037 (.078)	−.166 (−.078)	.633 (.680)	.036 (.117)	.253 (.374)
19	Start conversations unrelated to what I’m doing^a	.074 (.198)	−.071 (−.011)	.784 (.791)	.031 (.055)	−.018 (.173)
20	Play around with other things while doing my math homework^a	.063 (.189)	−.108 (−.032)	.763 (.797)	−.012 (.053)	.122 (.282)
21	Stop math homework repeatedly to find something to eat or drink^a	−.073 (.051)	.049 (.033)	.748 (.712)	−.044 (−.042)	−.102 (.056)
22	Stop math homework to send or receive “instant messaging”^a	.001 (.144)	.139 (.159)	.668 (.659)	.008 (.053)	−.068 (.123)

Note. The bolded pattern coefficients represent items considered to load on an appropriate factor. HMS = Homework Management Scale.

The item was reverse scored.

Based on the results of a study of U.S. high school students (Xu, 2008b), alpha reliability coefficient for scores on the 22-item HMS was .88. Its 95% CI was [.87, .89]. For the five subscales, the values were .75 (.72−.77) for Arranging Environment, .74 (.71−.77) for Managing Time, .74 (.72 −.77) for Handling Distraction, .83 (.81−.85) for Monitoring Motivation, and .80 (.78−.82) for Controlling Emotion.

The HMS was translated into Chinese (including the practice of back-translation) by a team of academics who were competent in both Chinese and English. The only adaptation made during the process is that we have specifically focused on math homework in the present study with high school students in China, instead of homework in general (as in the case with its original version of the HMS; Xu, 2008b).

Data Analysis

Data analyses were conducted in several steps. First, EFA is warranted, even when theoretical expectations are present regarding the number of factors (Henson & Roberts, 2006). This is a particular case for our study, as its structure has not been examined previously with high school students in China in the context of math homework.

The Chinese sample was first randomly split into two independent groups. For Group 1 (n = 442), principal components analysis using direct oblimin rotation (with delta = 0) was performed on the scores of the 22-item HMS. The decision about number of factors to retain was based on a combination of methods (e.g., eigenvalue > 1.0 and the scree plots) as well as conceptual meaningfulness of the rotated factors.

Second, the validity of a five-factor structure for Group 2 (n = 442) was tested using CFA. Specifically, the model hypothesized a priori that (a) responses to the HMS could be explained by the five factors labeled “environment,” “time,” “distraction,” “motivation,” and “emotion”; (b) each item would have a nonzero loading on each factor that it was designed to measure and zero loadings on all other factors; (c) the five factors were correlated; and (d) the error-uniqueness terms associated with the item measurements were uncorrelated.

Model fit was assessed by several indices. The ratio of χ² to its degree of freedom (χ²/df) with a range of less than 3.0 indicates an acceptable model fit (Carmines & McIver, 1981). The CFI values near 1.0 are optimal, with values greater than .90 indicating acceptable model fit (Kline, 2005). A value of .90 or greater was initially suggested as evidence of adequate fit. However, it was later suggested a value of .95 as a criterion for adequate fit (Hu & Bentler, 1999). Recently, the cutoff value of .95 is viewed as too restrictive (Byrne, 2008; Marsh, Hau, & Wen, 2004). Byrne (2008) suggests that CFI values in the range of .92 through .94 may be considered as reasonable indicators of good model fit, whereas Hair, Black, Babin, and Anderson (2010) recommend value of equal to or more than .90 to indicate an acceptable level of model fit. The RMSEA value less than .05 indicates a good fit, with values as high as .08 representing reasonable errors of approximation in the population (Byrne, 2008). The SRMR values less than .08 indicate a well-fitting model (Hu & Bentler, 1999).

Third, regarding validity evidence for the HMS scores, the participants were asked about their reasons for doing homework, based on the Homework Purpose Scale (HPS; Xu, 2010). The HPS consists of 15 items, using a 4-point format from 1 (strongly disagree) to 4 (strongly agree). It comprises three subscales: (a) Learning-Oriented Reasons (9-item subscale, for example, “doing homework helps you understand what’s going on in class”), (b) Adult-Oriented Reasons (3-item subscale, for example, “doing homework brings you family approval”), and (c) Peer-Oriented Reasons (3-item subscale, for example, “doing homework gives you opportunities to learn from classmates”). Based on the previous findings with U.S. students (Xu, 2010), alpha reliability coefficients for scores on these subscales were .79 (Peer-Oriented Reasons), .77 (Adult-Oriented Reasons), and .89 (Learning-Oriented Reasons). For the participants in this study, alpha reliability coefficients for scores on these subscales were .65, .84, and .89, respectively.

The students were also asked about their homework effort, adapted from the items used by Trautwein et al. (2006). The scale used in the present study includes four items, using a 4-point format from 1 (strongly disagree) to 4 (strongly agree). These items range from not copying from others to trying my best on math homework. Alpha reliability coefficient for scores on this scale for the present study was .80.

In addition to the above criterion measures (i.e., HPS and homework effort), the participants were asked, “Some students often complete math homework on time; others rarely do. How much of your assigned math homework do you usually complete?” Possible responses included 1 (none), 2 (some), 3 (about half), 4 (most), and 5 (all). They were further asked, “How often do you come to class without your math homework?” Possible responses included 1 (never), 2 (rarely), 3 (sometimes), 4 (often), and 5 (routinely). Finally, Pearson correlation coefficients were calculated (a) among the HMS and its subscales and each subscale of the HPS and (b) among the HMS and its subscales and homework behavior (i.e., homework effort, the amount of math homework completion, and the frequency of coming to class without math homework).

There were very few missing values for the 22-item HMS, ranging from 0.11% to 0.57% (with a mean of 0.32%). These missing values were imputed using the expectation-maximization.

Results

EFA

The Kaiser–Meyer–Olkin measure of sampling adequacy index for Group 1 (n = 442) was .823, indicating that the sample was appropriate for factor analysis. EFA yielded a five-factor solution that accounted for 60.06% of the variance in the HMS scores, and that was remarkably consistent with the previously validated five features of homework management for U.S. high school students (Xu, 2008b). All 22 items loaded substantially on the five factors that could be labeled as Arranging Environment, Managing Time, Handling Distraction, Monitoring Motivation, and Controlling Emotion. The factor pattern and structure coefficients are presented in Table 1.

CFA

CFA was used to test for the validity of the HMS structure described above with a sample of high school students in China (Group 2). An examination of the univariate sample statistics indicated that skewness and kurtosis values can be considered rather satisfactory—Only three items had kurtosis values larger than the absolute value 1 (−1.03 and −1.07 for Items 1 and 8, respectively) and only one item had a skewness value larger than the absolute value 1 (−1.66 for Item 5). However, the multivariate sample statistics were suggestive of non-normality in the sample, as evident by Mardia’s normalized estimate (44.91), which was greater than the cutoff point of 5.00 suggested by Bentler (2006). Thus, robust statistics (i.e., Satorra–Bentler), rather than regular statistics, were used to take into account some non-normality in the data.

Our initial testing of the hypothesized model for this sample yielded an adequate fit to the data as indicated by the following: S-Bχ²/df = 2.271; *CFI = .914; SRMR = .069; *RMSEA = .054; 90% CI = [.047, .060]. Similar to previous findings with U.S. high school students (Xu, 2008b), examination of modification indexes related to the data that identified two large correlated errors: one between Items 11 and 12 relating to monitoring motivation (good effort vs. good work) and another between Items 3 and 4 relating to arranging the environment (remove things from the table vs. make enough space to work). Given the substantive reasonableness of these parameters, a model was specified for Chinese students in which both parameters were freely estimated. The results indicated a good fit of the model to the data for high school students in China: S-Bχ²/df = 1.886; *CFI = .941; SRMR = .058; *RMSEA = .045; 90% CI = [.038, .052].

Concurrent Validity

Coefficient alpha for scores on the 22-item HMS for two groups combined (N = 884) was .84. Its 95% CI was [.83, .86]. For scores on the five subscales, the values ranged from .72 (Arranging the Environment) to .86 (Controlling Emotion). These reliability estimates are largely in the adequate to good range (Henson, 2001). These estimates, along with descriptive statistics and intercorrelations among the five subscales, are presented in Table 2.

Table 2.

Descriptive Statistics and Intercorrelations Among the HMS Subscales (N = 884).

Subscales	α (CI)	M	SD	1	2	3	4	5
1. Environment	.72 [.68, .74]	3.701	.846	—
2. Time	.77 [.74, .79]	2.968	.959	.334**	—
3. Distraction	.80 [.78, .82]	3.665	.834	.162**	.071*	—
4. Motivation	.82 [.80, .84]	2.900	.888	.208**	.389**	.072*	—
5. Emotion	.86 [.84, .87]	3.484	.827	.274**	.336**	.234**	.544**	—

Note. HMS = Homework Management Scale.

p < .05. **p < .01.

To examine the concurrent validity of the HMS, we examined the relationship between scores on the HMS and scores assessing homework purpose and behavior. As the significance students attach to academic tasks is critical for the efforts they contribute to the endeavor and the persistence they display (Eccles & Wigfield, 2002), as their views about homework play an important role on homework behavior (Cooper, Lindsay, Nye, & Greathouse, 1998; Warton, 2001; Xu, 2005), including homework management strategies that they use to aid homework completion (Xu, 2007; Xu & Corno, 2003), we hypothesized that the HMS and its five subscales would be positively related to homework purpose. As illustrated in Table 3, correlations coefficients among these variables were positive and statistically significant, with the exception of one coefficient between handling distraction and peer-oriented reasons. Specifically, the HMS and its five subscales were more strongly associated with learning-oriented reasons than adult-oriented or peer-oriented reasons. One possible explanation for no relationship between handling distraction and peer-oriented reasons is that those students with higher scores in peer-oriented reasons are more likely to interact with peers about homework tasks. As cooperative learning activities are likely to contain peer distraction (Corno, 2004), interaction with peers over homework may also lead them to engage in other social activities unrelated to the homework task at hand, thereby making them to take less initiatives to cope with homework distraction.

Table 3.

Pearson Correlations Between Homework Management, Homework Purpose, and Homework Behavior (N = 884).

Homework purpose and homework behavior	Environment	Time	Distraction	Motivation	Emotion	HMS
Homework purpose
Peer-oriented reasons	.125**	.237**	.023	.249**	.208**	.256**
Adult-oriented reasons	.089**	.220**	.121**	.270**	.171**	.268**
Learning-oriented reasons	.156**	.259**	.272**	.316**	.328**	.413**
Homework behavior
Homework effort	.198**	.235**	.358**	.248**	.407**	.451**
The amount of math homework students completed	.134**	.158**	.293**	.125**	.288**	.314**
The frequency of coming to class without math homework	−.092**	−.160**	−.241**	−.047	−.201**	−.235**

Note. HMS = Homework Management Scale.

p < .05. **p < .01.

We further examined the relationship between scores on the HMS and homework behavior. As expected, the HMS and its five subscales were positively associated with math homework effort and completion and negatively related to the frequency of coming to class without math homework (with the exception of monitoring motivation). Taken together, these correlations were of magnitude and direction consistent with theoretical expectations, thereby providing further support to the convergent validity of the HMS for high school students in China.

Discussion

The aim of this research was to adapt the HMS for high school students in China relating to their math homework and to examine its psychometric properties. Our findings revealed that the Chinese version of the HMS has shown adequate psychometric quality relating to measurement reliability and validity. Specifically, the findings from EFA and CFA lent empirical support to the existence of five separate, yet related, dimensions in math homework for high school students in China: Arranging the Environment, Managing Time, Handling Distraction, Monitoring Motivation, and Controlling Emotion. These findings are in line with previous findings with U.S. high school students (Xu, 2008b). Our findings further revealed that the HMS and its five subscales were positively related to homework purpose (peer-, adult-, and learning-oriented reasons) and desirable homework (i.e., homework effort and completion) and were negatively related to undesirable homework behavior (i.e., the frequency of coming to class without math homework), with the exception of 2 out of 36 possible correlations.

Taken together, and consistent with the previous findings that the HMS is factorially valid measure of homework management for U.S. high school students (Xu, 2008b), our study suggests that the HMS is applicable to high school students in China in the domain of math homework. Thus, the HMS seems to represent a useful, reliable, and valid means for assessing high school students’ homework management in the United States and China and across different subject areas. This is substantiated by relevant evidence from our study and the previous study (Xu, 2008b) which showed that for both students in the United States and China, in the context of homework in general and with math homework in particular: (a) the HMS comprised the same five-factor structure; and (b) reliability coefficients of the scale and its subscales were in the adequate to good range. In addition, in line with the previous finding with U.S. high school students that the HMS was positively related to homework completion, the present study revealed that the HMS and its subscales were positively related to homework utility value (i.e., homework purpose) and homework effort and was negatively related to undesirable homework behavior (i.e., failure to bring math homework to class) in the theoretically expected directions (e.g., Cooper et al., 1998; Eccles & Wigfield, 2002; Warton, 2001; Xu, 2005).

Along with the previous study on the HMS for U.S. high school students (Xu, 2008b), the present investigation has addressed an important gap in research on self-regulation (e.g., the need to assess similarities and differences between the use of self-regulation strategies in different settings; Boekaerts, Maes, & Karoly, 2005) and in research on homework (e.g., the need to study non-achievement-related effects of homework; Cooper et al., 2006). Thus, the findings regarding the HMS are likely to have substantial utility to researchers and practitioners in different countries who are interested in self-regulation, with homework management in particular. The HMS is likely to have substantial utility to those who are interested in the linkage between homework management and other important variables in the homework process, including homework interest (Warton, 2001; Xu, 2008a), homework completion (Cooper et al., 1998; Xu, 2011), and academic achievement (Cooper et al., 1998). This scale may also be readily used by teachers and parents as a cost-efficient assessment tool. For example, they may use the HMS to gain a better understanding of how students manage homework. As a result, they will be in a better position to provide more relevant support for students’ effort at homework management (e.g., managing homework time; Xu, Yuan, Xu, & Xu, 2014).

Regarding future research, it would be important to examine the validity of scores on the HMS in other countries and in other subject areas (e.g., science homework). Although our study revealed that the HMS and its subscales were positively related to homework purpose and desirable homework behavior and were negatively related to undesirable homework behavior, it was limited to self-reported measures. Thus, it would be informative to incorporate other measures of homework behavior (e.g., homework effort and completion rated by teachers or parents) in future validation studies of the HMS.

Another interesting line of research relating to the validity of the HMS would be to intervene those students with lower scores on certain subscales (e.g., Managing Time) to improve their homework management strategies, then assessing the influences on these and other subscales as well as other homework measures (e.g., Homework Completion) and subsequent academic achievement. Finally, as the demand associated with homework may evolve over time (e.g., online homework or group homework; Xu, Du, & Fan, 2013), it would important to refine and further test the validity of the HMS for it to be relevant for researchers and practitioners.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

References

Bentler

P. M.

(2006). EQS structural equations program manual. Encino, CA: Multivariate Software.

Boekaerts

Corno

(2005). Self-regulation in the classroom: A perspective on assessment and intervention. Applied Psychology: An International Review, 54, 199-231.

Boekaerts

Maes

Karoly

(2005). Self-regulation across domains of applied psychology: Is there an emerging consensus? Applied Psychology: An International Review, 54, 149-154.

Byrne

B. M.

(2008). Testing for multigroup equivalence of a measuring instrument: A walk through the process. Psicothema, 20, 872-882.

Carmines

E. G.

McIver

J. P.

(1981). Analyzing models with unobserved variables: Analysis of covariance structures. In Bohrnstedt

G. W.

Borgatta

E. F.

(Eds.), Social measurement: Current issues (pp. 65-115). Beverly Hills, CA: Sage.

Chen

Stevenson

H. W.

(1989). Homework: A cross-cultural examination. Child Development, 60, 551-561.

Chen

Uttal

D. H.

(1988). Cultural values, parents’ beliefs, and children’s achievement in the United States and China. Human Development, 31, 351-358.

Cooper

Lindsay

J. J.

Nye

Greathouse

(1998). Relationships among attitudes about homework, amount of homework assigned and completed, and student achievement. Journal of Educational Psychology, 90, 70-83.

Cooper

Robinson

J. C.

Patall

E. A.

(2006). Does homework improve academic achievement? A synthesis of research, 1987-2003. Review of Educational Research, 76, 1-62.

10.

Corno

(2000). Looking at homework differently. Elementary School Journal, 100, 529-548.

11.

Corno

(2004). Introduction to the special issuework habits and work styles: Volition in education. Teachers College Record, 106, 1669-1694.

12.

Corno

(2004). Doing homework as the job of childhood. Theory Into Practice, 43, 227-233.

13.

Eccles

J. S.

Wigfield

(2002). Motivational beliefs, values, and goals. Annual Review of Psychology, 53, 109-132.

14.

Hair

J. F.

Jr. Black

W. C.

Babin

B. J.

Anderson

R. E.

(2010). Multivariate data analysis (7th ed.). Upper Saddle River, NJ: Prentice Hall.

15.

Henson

R. K.

(2001). Understanding internal consistency reliability estimates: A conceptual primer on coefficient alpha. Measurement and Evaluation in Counseling and Development, 34, 177-189.

16.

Henson

R. K.

Roberts

J. K.

(2006). Use of exploratory factor analysis in published research: Common errors and some comment on improved practice. Educational and Psychological Measurement, 66, 393-416.

17.

Hong

Peng

Rowell

L. L.

(2009). Homework self-regulation: Grade, gender, and achievement-level differences. Learning and Individual Differences, 19, 269-276.

18.

Hong

Wan

Peng

(2011). Discrepancies between students’ and teachers’ perceptions of homework. Journal of Advanced Academics, 22, 280-308.

19.

L. T.

Bentler

P. M.

(1999). Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Structural Equation Modeling, 6, 1-55.

20.

Kline

R. B.

(2005). Principles and practice of structural equation modeling (2nd ed.). New York, NY: Guilford Press.

21.

Kouzma

N. M.

Kennedy

G. A.

(2002). Homework, stress, and mood disturbance in senior high school students. Psychological Reports, 91, 193-198.

22.

Kirkup

(2007). Gender and cultural differences in Interne use: A study of China and the UK. Computers & Education, 48, 301-317.

23.

Marsh

H. W.

Hau

K.-T.

Wen

(2004). In search of golden rules: Comment on hypothesis testing approaches to setting cutoff value for fit indexes and dangers on overgeneralizing Hu and Bentler’s findings. Structural Equation Modeling, 11, 320-341.

24.

Power

T. J.

Dombrowski

S. C.

Watkins

M. W.

Mautone

J. A.

Eagle

J. W.

(2007). Assessing children’s homework performance: Development of multi-dimensional, multi-informant rating scales. Journal of School Psychology, 45, 333-348.

25.

Trautwein

Ludtke

Schnyder

Niggli

(2006). Predicting homework effort: Support for a domain-specific, multilevel homework model. Journal of Educational Psychology, 98, 438-456.

26.

Verma

Sharma

Larson

R. W.

(2002). School stress in India: Effects on time and daily emotions. International Journal of Behavior Development, 26, 500-508.

27.

Wang

D. B.

(2004). Family background factors and mathematics success: A comparison of Chinese and US students. International Journal of Educational Research, 41, 40-54.

28.

Warton

P. M.

(2001). The forgotten voices in homework: Views of students. Educational Psychologist, 36, 155-165.

29.

(2004). Family help and homework management in urban and rural secondary schools. Teachers College Record, 106, 1786-1803.

30.

(2005). Purposes for doing homework reported by middle and high school students. Journal of Educational Research, 99, 46-55.

31.

(2007). Middle school homework management: More than just gender and family involvement. Educational Psychology, 27, 171-187.

32.

(2008a). Models of secondary students’ interest in homework: A multilevel analysis. American Educational Research Journal, 45, 1180-1205.

33.

(2008b). Validation of scores on the homework management scale for high school students. Educational and Psychological Measurement, 68, 304-324.

34.

(2010). Homework purpose scale for high school students: A validation study. Educational and Psychological Measurement, 70, 459-476.

35.

(2011). Homework completion at the secondary school level: A multilevel analysis. Journal of Educational Research, 104, 171-182.

36.

Corno

(1998). Case studies of families doing third-grade homework. Teachers College Record, 100, 402-436.

37.

Corno

(2003). Family help and homework management reported by middle school students. Elementary School Journal, 103, 503-518.

38.

Fan

(2013). “Finding our time”: Predicting students’ time management in online collaborative groupwork. Computers & Education, 69, 139-147.

39.

Yuan

(2014). Modeling students’ time management in math homework. Learning and Individual Differences, 34, 33-42.

40.

Zimmerman

B. J.

Kitsantas

(2005). Homework practices and academic achievement: The mediating role of self-efficacy and perceived responsibility beliefs. Contemporary Educational Psychology, 30, 397-417.