A Validation of the Family Involvement Questionnaire-High School Version

of the Dissertation The purpose of this study was to validate the Family Involvement Questionnaire (FIQ) for use in high school settings (9 th -12 th grade). After the FIQ was redesigned for use in the high school setting, 517 parents completed the questionnaire online. Internal consistency for the 40-item questionnaire was high (α = 0.93). A confirmatory factor analysis failed to substantiate the FIQ-HS to the elementary version from which it was adapted. However, an exploratory factor analysis yielded three factors consistent with the FIQ-E. Family demographics were compared to participants’ responses, and significant effects for students’ school and special education status were found.


Introduction
The discussion of family involvement in children's schooling is not new. Buchanan, Hansen, and Quilling (1969) conducted one of the first studies examining the relationship between family involvement and student performance in the late 1960's.
Educational laws in the United States have mandated the involvement of parents in their children's education since the 1970's (EAHCA, 1975;IDEA 2004;NCLB, 2001). The first mandate specifically requiring parent involvement in schools was in 1975 with the The factors and benefits of family involvement have been examined in over forty years of research, and multiple meta-analyses have been conducted to synthesize the data (Fan & Chen, 2001;Jeynes, 2005Jeynes, , 2007Jeynes, , & 2012Mattingly, Prislin, McKenzie, Rodriguez, & Kayzar, 2002). This body of research provides a firm basis to conclude that family involvement can have a positive effect on school children's achievement as measured by grades, standardized test scores, school enrollment, and high school graduation rates (Catsambis, 1988;Fan & Chen, 2001;Hill & Taylor, 2004;Jeynes, 2005).
In 2006, Appleseed completed an investigation of educational laws, policies, and practices and their impact on family involvement practices in schools. In this investigation, researchers interviewed educational leaders, community organizations, and school district staff. They also conducted parent focus groups and reviewed research on family involvement. Overall, the researchers felt that schools do not accept family involvement as a main strategy for making academic gains, and current federal laws requiring parent involvement are not being followed by districts, likely because of a lack of understanding and support. Appleseed's report concluded that: Parental involvement is not uniformly valued by school leaders as a key accountability strategy. (p.2) Differences in family involvement between primary and secondary grade levels are apparent in both research and practice. Case in point, parent attendance rates to school functions typically decrease as students enter secondary grade levels. Researchers at the National Center for Education Statistics (2013) found that 89 percent of parents of kindergartners through fifth graders regularly attended parent-teacher conferences. That statistic dropped to 71 percent for parents of sixth through eighth graders, and dropped down to 57 percent attendance rate for parents of ninth through twelfth grade students.
Other differences between primary and secondary grades were found in this national household education survey, including differences in parent participation in school, educational habits in the home, and parents' school satisfaction levels (National Center for Education Statistics, 2013). The rate at which parents volunteered at school or served on school committees dropped significantly from primary grades to secondary grade levels. However, more parents reported meeting with a school counselor when their child was in secondary school versus when their child was in primary school.
This survey also asked parents about their school-related parenting behaviors in the home setting and their satisfaction levels (National Center for Education Statistics, 2013).
Parents reported high rates of ensuring their child had a place in their home for them to complete homework at all grade levels. However, parents of secondary students reported much lower rates for checking that their homework is complete than parents of primary students. Parent satisfaction with their child's school and teachers also yielded significant differences. More parents reported that they were very satisfied with their child's teacher(s) if their child was in a primary grade, than if their child was in a secondary grade. More parents also reported that they were very satisfied with the way in which school staff interacts with parents when their child was in a primary grade, versus a secondary grade.

Statement of Problem
Literature on the effects of family involvement in school settings has focused primarily on early childhood and primary school settings (Fantuzzo, Tighe, & Childs, 2000;Manz, Fantuzzo, & Power, 2004;Waanders, Mendez, & Downer, 2007). This literature has established a firm base of knowledge that family involvement in schooling can positively impact students' school performance. However, significant differences in family involvement practices and behavior between primary and secondary grade levels are evident (National Center for Education Statistics, 2013). As educational laws that mandate the involvement of families in the educational process cover all grade levels, it is important to have valid measures for assessing involvement in the lesser-studied secondary grade level. The current study will seek to validate the Family Involvement Questionnaire for use in the high school level (9 th through 12 th grades).

Research Questions
In order to validate the Family Involvement Questionnaire (FIQ) for the high school level (FIQ-HS), a sample will be collected and statistical analyses will be conducted to establish the reliability and validity of the instrument. Additional correlation analyses will be conducted to examine demographic data with the instrument. By conducting these analyses, we hope to answer three main research questions: 1) Does the FIQ-HS demonstrate internal consistency?
2) What factors are associated with family involvement in high school settings?
3) Are the factors found in the FIQ-HS consistent with the ones found in previous FIQ versions, including Home-School Communication, Home-Based Activities, and School-Based Activities factors?
Previous research in family involvement has not examined the factors of involvement in the high school setting through the use of this questionnaire. Because of this, expected outcomes for this study are difficult to hypothesize. It is expected that there will be differences between the elementary and high school versions of the FIQ, as each setting has its own unique family involvement structure and expectations. Although differences between elementary and high school settings are evident, general factor patterns are expected to be consistent between the two settings.

Significance of the Study
The results of this study will add to the literature base on family involvement in schools by assessing involvement within a setting that previous research has not assessed in such a way. The data collected from this study can also be used within the individual schools, and others of similar structure, to help develop interventions targeted towards increasing specific aspects of family involvement. As previous literature has established, students of families that are regularly engaged in school activities have higher school performance (Fan & Chen, 2001;Jeynes, 2005Jeynes, , 2007Jeynes, , & 2012Mattingly et al., 2002). As family involvement practices can be relatively inexpensive or free for schools to implement, programs targeting family involvement should be in place at every school.
The United States' educational laws have further reinforced the importance of family involvement in schools (IDEA, 2004;NCLB, 2001). IDEA requires each state to provide the federal government with data on parent involvement facilitated by schools, but gives little guidance on how to collect such data. Elbaum (2014) found high variability between the states in their reported data, likely owing to the different measures states chose to use when collecting their data. Due to the inconsistencies between the states, comparing data from different states or synthesizing data becomes inaccurate.
With the results of this study, participating schools will be able to review their current overall level of family involvement, and then target specific factors or aspects of involvement they wish to increase. Schools may also use this assessment at a later date to compare any possible changes in the family involvement levels once programs or initiatives have been put into place.
Another function of this study is to extend the FIQ, so the measure can be used to assess family involvement in high school-aged students' families. The majority of research on family involvement in schools has focused on early childhood and primary school grade levels. It is likely that the components of family involvement change as students age into adolescence. By extending the FIQ questionnaire to the high school level, more research can be conducted in this setting through the use of this simple questionnaire.

Scope of the Study
Data will be collected using the FIQ from five high schools in Minnesota, United States. The data obtained from each school will be compiled and analyzed as a whole for the purpose of this study. Family involvement results of individual school sites will then be analyzed to make recommendations specific to each school site.
To extend the FIQ to the high school level, items from the FIQ-E will initially be analyzed for their appropriateness and relativity to secondary-aged students and their families. Once data from the high school version of the FIQ has been collected, analyses including Cronbach's alpha and factor analyses will be conducted on the questionnaires completed to establish the reliability and validity of the instrument in the high school setting. Analyses will be completed to examine any possible relationships between participants' responses on the FIQ and their responses to the demographic questionnaire.
Factor loadings and structures from the factor analysis will also be compared with previous versions of the FIQ. Sample demographics data will also be calculated and reported.

Limitations of the Study
In this study, the extension of the FIQ to the high school level is limited by geographic region and small sample size. The geographic region of this sample will be confined to the state of Minnesota. The sampling pool for this portion of the study is intended to consist of approximately 500 parents. While this sample will be large enough for an initial factor analysis to establish the FIQ within the high school setting, results should not be generalized based on the small sample and confined setting characteristics.

Definition of Terms
Family Involvement: The term family involvement will be used to describe the participation of parents in the education of their children. This includes meaningful communication with school staff and performing behaviors that are related to their child's learning within the school, home, or community settings (United State Department of Education, 2004). Epstein (1992) defined family involvement into six specific components; these components are associated with the development of the original FIQ: 1) Assisting parents in child-rearing skills, 2) School-parent communication, 3) Involving parents in school volunteer opportunities, 4) Involving parents in homebased learning, 5) Involving parents in school decision-making, 6) Involving parents in school-community collaborations. (pp. 1142-1150) Parent: The term parent is defined as a student's biological parents, legal guardians, or primary caregivers. For the purpose of this study, parents are further defined as being at least 18 years old.

Student:
The term student is defined as a school-aged (5-21 years old) person, who is enrolled full-time in a high school system (grades 9 th through 12 th ).

Overview of Study
This study is organized into five sections. Section one is an introduction to the study that outlines the framework in which the study was developed. The importance of family involvement in schools is established through a literature review in Section two.
The literature review also discusses the current research on family involvement. The methods used to collect and analyze the data are outlined in Section three, and the results derived from the data collection are in Section four. Section five is an in-depth discussion of the results found in the comparative analyses and factor analysis of the FIQ-HS.
Implications of this study as well as recommendations for future research are also discussed.

Overview
The notion that family involvement has a positive impact on student school performance is easy to accept. Not necessarily because of the vast amount of empirical research that has been conducted and published in peer-reviewed journals, but because on a basic level, it appears intuitive. The way in which researchers measure family involvement varies depending on their interpretation and definition of this construct.
Undoubtedly, family involvement is a multifaceted construct, and therefore will have varying definitions between researchers. Differences between primary and secondary aged students and their families further confound the definition and measure of this construct.
Quantitative research on the effects of family involvement on student school performance officially began in the late 1960's. Buchanan, Hansen, and Quilling (1969) examined the relationship between the frequency of contact between students' home and school environments, and their performance in math. Their study was the first published work to examine family involvement and academic performance in a quantitative manner through an experimental group design. Although their study yielded insignificant results for the effects on the students' math performance over the 12-week study schedule, numerous other studies examining family factors would follow over the next forty years.

Researchers have had difficulty defining what is family involvement in schools.
Different school systems promote different types of involvement programs for families of their students. Schools have the choice in the type of programs they choose to implement or support. The way in which schools are selecting such programs likely range from empirical support, to easiness and cost, to continuing with what has historically been done in their school system. Each family also has their own opinions on what behaviors of family involvement are important to them. These opinions are likely shaped by their own experiences, knowledge, and cultural background.
Huntsinger and Jose (2009) examined parental involvement behaviors in two United States cultures: European Americans and Immigrant Chinese Americans parents.
They found a distinct difference in behavior patterns between the two cultures. European American parents were more involved with activities at their child's school. However, Chinese Americans were more involved in educational training at home. Another interesting difference between the two cultures was that European American families preferred a tradition grading system (e.g., letter grades, plus/minus) for communicating academic progress. Chinese American families preferred more detailed communication with teachers, including specific concepts their child has mastered or still needs to learn.
This study demonstrated how cultural differences in family involvement are not only evident between countries, but are seen within one as well.
Using Bronfenbrenner's (1992) ecological systems theory, it is easy to explain the influences of both family and school on a child and why collaboration between them is important. Both of these institutions fall within a child's microsystem and have direct impact on their development. For children, school is their main focus, much like a job is for an adult. The child's family is designed to be a consistent support of basic needs and foster growth and development. These two institutions likely have the greatest direct impact on a child, so bridging a partnership between the two is important. When problems, trauma, or change occurs in one of these settings it will likely affect the child's behavior in the other. The importance of this partnership is seen with Bronfenbrenner establishing the mesosystem within his model depicting these interconnections between the child's microsystems. Christenson (2004) supports family involvement practices in schools as they can create a synergistic relationship between these two institutions that are central to a child's development. She discusses several barriers as to why this partnership is not always achieved, which include not only structural or practical barriers, but psychological ones as well. Christenson argues that school psychologists are aptly trained to have a leading role in bridging home-school partnerships at a systems level. Pelco, Ries, Jacobson, and Melka (2000) conducted a national survey of school psychologists to examine their views and practices of family involvement in schools.
Overall, school psychologists supported parent involvement in education as a way to increase student success. They viewed their roles in consulting with families, teaching parenting skills, and facilitating conferences as very important, and a majority of respondents reported regularly participating in these activities.

Components of Family Involvement
In 1987, Epstein identified four types of parental involvement in school: Basic Obligations, School-Home Communications, Parental Involvement at School, and Parent Involvement in Learning Activities at Home. In 1992, Epstein further defined her types of parental involvement to include six different levels of involvement: 1) Assisting parents in child-rearing skills, 2) School-parent communication, 3) Involving parents in school volunteer opportunities, 4) Involving parents in home-based learning, 5) Involving parents in school decision-making, and 6) Involving parents in school-community collaborations.
In 2002, Epstein et al. simplified her six types of involvement into: parenting, communicating, volunteering, learning at home, decision-making, and collaborating with the community. In the parenting type of involvement, schools assist families by teaching parenting skills and providing support to families. Parents in turn are expected to share with the school their family backgrounds, values, and goals. The communicating type involves schools communicating with families about events, programs, and their individual child's progress. Epstein and colleagues stressed that schools should strive to create a two-way communication channel were the parent feels comfortable contacting the school with news or concerns. In the volunteering type, schools seek to recruit families as volunteers in a manner that supports students. In learning at home, schools strive to extend the child's learning to the home setting by having family members support the student through learning activities. In decision-making, families are included in school decisions through committees, action teams, or other organizations. In collaborating with the community, schools aid families by connecting them with community resources or services, as well as organizations, businesses, or post-secondary education.
Based on the description of these levels, Epstein takes the viewpoint of the school system. She appears to believe that it is the responsibility of the school system to facilitate and encourage family involvement. Current United States educational law, such as No Child Left Behind Act (NCLB, 2001), would support that the burden of facilitating and promoting family involvement falls onto the school system. Schools receiving Title 1 funding are also required to spend a portion of the funds on organizing and facilitating parent participation programs. Some researchers feel that Epstein's levels of involvement may be too simplistic to capture the true nature of family involvement (Jeynes, 2012). In 1995, Hoover-Dempsey and Sandler established their own theoretical framework for parental involvement. Since then, they have continued to rework their model with other researchers. In 2005, they presented a revised model, with five levels of involvement seen in Table 1. Smith and colleagues (2011) found that the biggest difference seen in Epstein's model in their charter schools was at level five, participation in school decision-making.
In some cases parents elect the school's governing board, and in others parents make up the governing board. The researchers found that administrators lacked confidence in situations when they needed to involve parents, possibly because of their lack of experience with working in parents in such a way in an educational system, or because of the power the parents held in these alternative school settings.
Specific components of family involvement were assessed in Jeynes ' (2005) meta-analysis. Parental expectations, parental reading, parental style, and specific parental involvement were found to significantly effect school performance. Parental expectations and style were found to have the strongest effect sizes, suggesting that specific behaviors were less influential. Checking homework, attending school functions, and establishing household rules did not produce significant effect sizes on school performance. The significant effect parental expectations and parenting style have on student outcomes outline the importance of the family's atmosphere and dynamics.
Not all family involvement practices are equal. Pomerantz, Moorman, and Litwack (2007) argued that more importance should be placed on the exact nature and way in which parents are being involved in their child's schooling. They concluded that when parent involvement was controlling, person focused, and was accompanied by a negative affect or negative beliefs it has a negative impact on student outcomes. Whereas when involvement was autonomy supportive, process focused, and associated with positive affect and beliefs, the parents' behaviors had a positive impact on student outcomes. Cox (2005) reviewed eighteen studies examining the effects of home-school collaboration interventions on student outcomes. She found that overall these interventions increased students' academic achievement and improved student behavior.
The interventions with the greatest effect sizes involved parents and teachers coimplementing an intervention and routinely exchanging information.

Effects and Associations of Family Involvement
The effects of family involvement on students' school performance have been well studied over the past four decades. A current search on the Education Resources Information Center (ERIC) database of, "parent, involvement, and schools," will yield over 11,600 journal articles, papers, and books. Multiple meta-analyses have been conducted using published studies to synthesize findings (Fan & Chen, 2001;Jeynes, 2005Jeynes, , 2007Jeynes, , & 2012Mattingly et al., 2002). Fan and Chen (2001) completed one of the first meta-analyses in this area. They narrowed down their initial pool of published studies and articles from over 2,000 to 25 empirical studies that met their inclusion criteria. When reviewing the literature, they found that the effects of family involvement were inconsistent between studies. Fan and Chen found that studies in which the area of academic achievement was assessed by a more global indicator, such as GPA, had greater effect sizes than studies in which achievement was measured by specific subject grades or single test scores.
Another interesting finding in Fan and Chen's (2001) meta-analysis, was that parental involvement when measured by direct supervision had a very low effect size on school performance. The largest effect sizes for parental involvement were found when it was measured by parent's expectations or aspirations for their child's performance. These differences in effect sizes for different parenting behaviors highlight the importance of determining what programs should be targeted by schools. Fan and Chen (2001) calculated an overall correlation of r = .25, which is considered small, but meaningful. The authors noted that they believed this is a low estimate of the true effect size of parental involvement on school performance. They felt that studies in which a global indicator of school performance (e.g., GPA) was used as a measure were more accurate, and calculated such studies as having a higher correlation to family involvement (r = .33). Studies in which single subject grades were used as a measure had a smaller correlation (Science r = .15, Math r = .18, Reading r = .18).
One year later, Mattingly, Prislin, McKenzie, Rodriguez, and Kayzar (2002) reviewed 41 studies that targeted parent involvement programs' relationship to student In 2012, Jeynes conducted another meta-analysis, this time focusing on parental involvement programs used in urban schools. Overall, he found that such programs were associated with a .30 standard deviation increase in school performance. Specific programs with significant effect sizes included shared reading, emphasized partnership, checking homework, and communication between parents and teachers. Shared reading programs had the greatest effect size (.51). These were programs that encourage parents to read with their children, either using materials provided or recommended by the school, or materials chosen by parents. Programs with low effect sizes included English as a Second Language programs targeted towards parents and Head Start programs that emphasized parent participation.
Jeynes (2012) also noted that the majority of studies he reviewed were conducted in primary school grades. He suggests that this shows the value schools place on parental involvement in lower grades compared to the secondary level. Jeynes also suggests that schools find it easier to involve parents of younger children, making parent involvement programs utilized more than they may otherwise be at a secondary level. Although, the similarity in effect sizes Jeynes calculated for primary students (.29) and secondary students (.35) would suggest that parent involvement programs can have similar effects with older students.
El Nokali, Bachman, and Votruba-Drzal (2010) reviewed data from the National Institute of Child Health and Human Development's study on early childcare and youth development to assess elementary children's academic and social development in relation to their family's involvement in school related activities. They found that increases in parent involvement were associated with decreases in behavior problems and increases in children's social skills. However, increased levels of parent involvement were not associated with an increase in academic achievement.
The Center for Research on the Education of Students Placed At-Risk sponsored a longitudinal study investigating the effects of family involvement on at-risk secondary students' school performance (Catsambis, 1988). The report found that high educational expectations, regular encouragement, and behaviors that create or enhance learning opportunities were associated with positive effects on twelfth grade students' achievement, as measured by course credits and placement. Variables including socioeconomic status and ethnicity did not significantly interfere with the positive effects seen in families with high expectations for their students' educational performance.
Jeynes (2007) completed another meta-analysis focusing on urban secondary school-aged students. After reviewing 52 studies that included middle and high school students' families, he found that overall parental involvement was associated with a .50 to .55 increase of a standard deviation in performance on academic measures. These increases were lower than the results found in Jeynes' (2005) other meta-analysis focusing on elementary-aged students, which showed an overall increase of .70 to .75 of a standard deviation. Although a lesser effect was found within studies of secondary students, a meaningful increase in academic performance was still evident.
The evidence of positive effects connected to family involvement is present in the literature base. So, how can schools increase family involvement? Some of the significant influencers of parental involvement are out of schools' control, but schools can manipulate others. If targeting school-based involvement, research suggests that parental role activity beliefs, parental self-efficacy, teacher invitations, student invitations, and parents' reports of time and energy for involvement are the greatest predictors of engaging in school-based involvement (Green et al., 2007). Schools can attempt to increase their parent involvement at school by having teachers and students make specific requests or invitations to parents to participate, and by describing to the parents exactly what would be expected of them during the involvement. General invitations from the school and parents skills or knowledge were not important predictors of parent involvement in school-based activities.

Adolescents and Secondary Settings
Research in the area of family involvement practices in schools has focused primarily on early childhood and elementary settings (Fantuzzo et al., 2000;Manz et al., 2004;Waanders et al., 2007). This focus may be because of the natural shift that occurs in family involvement practices as students age into adolescence, generally becoming less hands-on and more supportive in nature. However, family involvement practices have still been shown to significantly increase student outcomes at the secondary school level Jeynes, 2007;Spera, 2005).
In 2005, Spera published a review of family involvement practices in adolescence and their relationship to school achievement. He concluded that family involvement is a strong predictor of adolescent achievement, and that family involvement declines in middle and high school. Wentzel and Battle (2001) saw this decrease in family involvement as a natural progression, "a hallmark of adolescent development is gaining emotional and psychological independence from family" (p. 95). Suggesting that as students age into adolescence, the way in which families are involved in their schooling needs to change from direct involvement to a more supportive role in the home setting. Deslandes and Cloutier (2002) surveyed adolescents on their views of parent involvement in their schooling. Overall, they found that the adolescents supported the majority of family involvement activities, with females indicating a higher degree of support than males. The activities in which the adolescents agreed most to included: asking parents for ideas on projects, having their parent share stories from when they were a teenager, and showing their parent what they learned or completed. The two activities in which the adolescents were not supportive of were chaperoning school trips and visiting their classes. This implies that the participants preferred that their parents were not physically present in their school, but instead provide support in the home setting.
Developmentally, the decrease in parental involvement in schools as students age is logical. As students mature, they are expected to become more self-reliant and independent from their families. Izzo, Weissberg, Kasprow, and Fendrich (1999) conducted a longitudinal study to examine parent involvement in first through third graders. They found that parent-teacher interactions and parent involvement at school decreased at each grade level. Green and colleagues (2007) found a similar decline from first through sixth grade for both school-and home-based involvement. Spera (2005) showed that this trend continues through secondary grades. However, family involvement can still have a positive effect on secondary students' school performance. As children mature, the way in which families are involved needs to change Jeynes, 2007;Spera, 2005).
By comparison family involvement in the early-childhood and elementary levels is expected, fun, and easily understood by the parent and school. Eccles and Harold (1993) discuss creating meaningful and age-appropriate opportunities for family involvement in secondary schools. They suggest that schools have parents in important roles or on committees, begin communicating with families prior to the secondary transition, and create multiple opportunities for parents to support their child's schooling in the school and home settings.
Adams and Christenson (2000) surveyed parents and teachers about their level of trust in family-school relationships. Overall, parents reported higher levels of trust than teachers. They found that both parents and teachers reported higher levels of trust in elementary grades than secondary grades. Parents indicated that improving communication between school and family would most likely to increase family-school trust, while teachers indicated that demonstrating dedication to education and having a positive academic environment would increase the trust level most. Wheeler (1992) argued that family involvement in secondary schools is necessary for students to develop into successful adults. She points out the struggles in accomplishing this as adolescents tend to distance themselves from their families, and as the amount of students each teacher works with grows considerably. She recommends creating activities that are age appropriate, increasing home-school communication and home-based activities, and using advisors or other people that can be the main school contact for a family. She also stresses that schools should make their first contact positive, continue frequent communication, find positives to share with the family, ask for parent suggestions, encourage parents to visit the school, and be specific when asking for parent help.
Hornby and Witte (2010)  academic achievement, and post-secondary aspirations. They found that parental involvement continues to be a strong predictor of student achievement throughout secondary school. Overall, the researchers found that parent involvement in seventh grade was negatively correlated to school behavior problems in eighth grade, and positively correlated to post-secondary aspirations in eleventh grade. Differences across family socio-economic status were also found. Parents with lower education levels who were more involved in their child's schooling were associated with increased student aspirations for post-secondary schooling and career. However, increased parent involvement for these families was not associated with improved school behavior or achievement. Parents with higher education levels who were more involved were more likely to have students with fewer school behavior problems, and higher school achievement and post-secondary aspirations.
Hill and Tyson (2009) examined 50 studies to determine which types of family involvement are more strongly associated to school achievement in middle school students. In this meta-analysis, they found that overall family involvement was positively correlated to students' academic achievement. However, with great variability in the included studies correlations (-.49 to .73), their average weighted correlation of r = .18 fell within the weak range. This is likely because of the varying degrees in which the included studies chose to measure family involvement and academic achievement. Hill and Tyson determined that involvement that utilized academic socialization techniques had the greatest impact on students' achievement. Suggesting that parent involvement should focus on goal setting, discussing the purpose of education, and teaching adolescents strategies to use when making decision independently. Family involvement that was school-based, such as attending school events or volunteering at the school, was also correlated to higher achievement levels. Involvement that included providing assistance, checking, or supervising students' academic work was not significantly associated with student achievement.

Family Involvement Questionnaire: Early Childhood
Fantuzzo, Tighe, and Childs (2000)  The validation of this short form of the FIQ created a cost-effective measure that is less burdensome for parents to complete and researchers to analyze, but continues to be psychometrically sound.

Family Involvement Questionnaire: Elementary
In 2004

Purpose of Present Study
The purpose of the present study was to validate the FIQ for use in high school settings by demonstrating reliability and validity of the instrument within a high school parent sample. This instrument was previously been validated in early-childhood and elementary settings, but had yet to be used with families of high school students. The Additional analyses to examine correlations between participants' FIQ-HS responses and family characteristics were also conducted. In addition to analyzing the samples data as a whole, individual analyses were conducted for each participating school site to provide specific feedback and recommendations.

Participants
The participants were parents of high school students whose teenager was currently enrolled in ninth through twelfth grade. Five hundred and seventeen parents were recruited from five high schools in southern Minnesota, United States (Table 2).
They were 18 years of age or older, with a mean age of 45 (M = 44.72, SD = 5.25). Table   3 shows the indicated relationship of the participant to the high school student. Mothers were the predominant responders (79.70%) to the survey, with fathers being the second most likely to respond (17%). The majority (96.10%) of participants indicated their family ethnicity as Caucasian or White (Table 4). A majority (87.23%) of participants indicated that their high school student was not receiving special education services (Table 5).   Potential participants were recruited through email to participate in this study.
Participants were directed to an informed consent webpage, which they are expected to read before they completed the study materials. Anonymous informed consent was utilized because the research involved little risk and included no procedures for which written consent is normally required. Participants' informed consent was implied when completed survey materials were returned to the researchers via an online survey system.
Participants were not asked to provide any individually identifiable information (e.g., name, birthday, address) in the course of participating in this study. The FIQ-E was originally used with families of students in first through fifth grade. The items on the FIQ-E were examined to determine if they were appropriate for high school aged students in grades ninth through twelve. Researcher item examination was used to determine what items from the FIQ-E were applicable to high school aged students and what new items needed to be added to capture any unique family involvement aspect only seen at the high school level.

Measures
In the adaptation of the FIQ-HS 34 items were kept from the FIQ-E, and 11 items were removed because the behavior was not applicable to parents of high school students.
Examples of removed items included, "I read with my child," and "I go on class trips with my child." Six new items were created for the FIQ-HS (Appendix A; items 10, 12, 13, 28, 30, and 34). These new items reflected transition related parenting behaviors including preparing their teenager for post-secondary education, employment, and independent living.
Additional changes were made to items on the FIQ-HS to make the measure more appropriate for the high school level. These changes included the word "teacher" being pluralized to address the fact that high school students typically have multiple teachers at any one point in time, and changing the word "child" to "teen" in the items. Another significant change made to the FIQ-HS was that a directive was added to the beginning of the questionnaire asking parents to consider the multitude of educators that may not necessarily have the title of "teacher," but serve as a school support for them or their child. Parents who also have younger children who were not yet in high school were also asked to only consider their 9th through 12 th grade child(ren) when responding to items.
When the FIQ-E was developed for use with parents of children in primary school, three distinct factors emerged: Home-Based Involvement, Home-School Communication, and School-Based Involvement. Using these factors the researchers divided the 46 items into three scales representing each factor. It was anticipated that similar factor structures would emerge in the FIQ-HS, and the three scales could be established.
Demographic questionnaire: Demographic variables were gathered to gain an understanding of the sample used in this study (Appendix C). Specifically, items addressed participants' relationship to the student, age, ethnicity, and their child's special education status and school of attendance. The demographic questionnaire was completed by a primary caregiver of a high school student and took approximately 5 minutes to complete.

Procedures
To recruit potential participants, permission was first sought from individual school sites (Appendix D). High schools were recruited for their participation by contacting designated administrators, superintendents or principals. School consent was necessary to obtain contact information for high school parents. Once school consent was obtained, parents were then recruited to take part in the study. Five high schools in rural Minnesota signed permission and completed participated in this study.
School administrators were given the option of having parents contacted through postal mailings or email. All participating schools selected to have their parents contacted through email. School administrators were also given the option of releasing parent email addresses to the researchers and having them contact parents, or having the researchers provide the survey link to the school and they could disperse the link directly to parents.
All participating schools choose to disperse the survey link directly to their parents. No parent contact data was released to the researchers. An example email to parents was provided to school administrators (Appendix E), they were able to use this example, modify it, or create their own message. All participating schools choose to use the example email provided.
When participants were contacted they received an email from their respective school administrator briefly explaining the study and providing the survey ink.
Participants who selected the link were taken to an online survey system (Qualtrics).
They were initially directed to the informed consent webpage (Appendix F). Informed consent was established when a participant selected the "Yes, I agree to participate in this study," button at the bottom of the informed consent webpage and completed the FIQ-HS and demographic questionnaire. By submitting the completed questionnaires the participant also indicated that they were at least 18 years of age and a parent of a high school student.
Participating parents completed the FIQ-HS and demographic questionnaire at a location of their choice, likely in their own home or place of work. Participants completed the questionnaires through a secure online survey website and their responses were stored in an online database that could only be accessed by the researchers.
Approximately two weeks after school administrators sent out their initial email to their parents, researchers informed each school of the number of completed surveys for their school. At this time, researchers recommended sending out one follow-up email to parents reminding them of the survey (Appendix E). All five participating schools sent out a follow-up email to parents.
After data collection was completed, feedback reports for each participating school were created. These feedback reports included data extracted from their own parents as well as data from the whole sample for comparison purposes (Appendix G).
Data provided in the reports included: school demographics, performance by scales for school and sample, individual item mean and distribution by school and sample, and a strengths and weaknesses report. These feedback reports were distributed to school officials approximately one month after all data collection was completed.

Results
Data was analyzed in three ways. First, to establish the internal consistency of the FIQ-HS a Cronbach's alpha reliability analysis was conducted. Second, to establish the construct validity of the instrument multiple factor analyses were performed. Identified factors were examined for internal consistency and then compared to the FIQ-E. Lastly, relationships between family characteristics identified in the demographic questionnaire and participants' responses to the FIQ-HS were analyzed using a MANOVA to determine what family characteristics were associated with more or less involvement.

Internal Consistency
A Cronbach's alpha was calculated on the 40 items within the FIQ-HS (Cronbach, 1951). The FIQ-HS yielded high internal consistency with a coefficient of 0.93.

Confirmatory factor analysis
A confirmatory factor analysis was conducted to examine the structural validity of the FIQ-HS across three constructs identified in the FIQ-E (Manz et al., 2004). Thirtyfour items on the FIQ-HS were assigned to one of the three factors identified in the Manz et al. study. These 34 items were ones that were consistent with the FIQ-E version, with only minor wording changes made for the high school population. The confirmatory analysis indicated that the three-factor structure found in the FIQ-E is not applicable to the FIQ-HS using the current sample.

Exploratory factor analysis
Exploratory factor analyses were completed to identify constructs in the FIQ-HS.
An orthogonal (varimax) rotation was conducted first, yielding three factors across 25 items. An oblique (promax) rotation was then conducted on the theoretical basis that the three-factor structure identified in the FIQ-E could have relationships with one another.
By conducting the oblique rotation it allows for a small degree of correlation between factors, such as the home-school communication and school-based activities factors that were predicted. A three-factor solution was supported by both orthogonal and oblique rotations, with items loading consistently on the three factors across both methods. As the orthogonal method was used in previous validation studies of the FIQ results from this rotation were used for data reporting and interpretation.
The factor structures were examined using the criteria established by McDermott (1993). Factors with eigenvalues of less than 1 were eliminated, factors that accounted for less than 5 percent of the total variance were eliminated, and factors with unacceptable internal consistency (α < .70) were also eliminated. After this process, a three-factor structure was supported by the exploratory factor analysis. The following dimensions were produced across 25 items: home-school communication, home-based activities, and school-based activities. The internal consistency of each factor was good to acceptable, with Cronbach's alpha coefficients of 0.89, 0.71, and 0.77, respectively (Cronbach, 1951). In total, the three factors account for 31.67% of the variance, with the home-school communication factor accounting for 14.19%, the home-based activities factor accounting for 10.27% and the school-based involvement factor accounting for 7.20% of the variance.
The three factors item content and factor loadings are presented in Tables 6, 7, and 8. Items with factor loadings of less than 0.40 were deemed non-loading based on Stevens (2002) recommendation, however for future research these items and loadings are included with their anticipated factor.
The home-school communication factor is comprised of 11 items reflecting various forms of contact parents might have with school staff, including communication behaviors such as talking with teachers about difficulties at school, accomplishments, and policies, and contacting the school for information ( Table 6). The home-based involvement factor is comprised of 9 items and includes activities parents perform outside of school that promote learning, such as talking with their teenager about careers and schooling, and helping their teenager with homework ( Table 7). The school-based involvement activities factor is comprised of 4 items that reflect parent behavior in the school setting, such as volunteering, and participating in family social activities at school or school fundraising activities (Table 8).  In total, 24 of the 40 items on the FIQ-HS were identified across the three factors, with factor loadings of 0.40 or greater. If the factor loading threshold was set at 0.30 or greater, 29 items would load onto the three factors. There were 16 items that did not load onto any factor or did load with a factor loading of less than 0.40.

FIQ-E and FIQ-HS factor analysis comparison
Although the confirmatory factor analysis indicated the factors of the FIQ-E were inconsistent with the FIQ-HS, results of the exploratory factor analysis revealed similarities. First, the three constructs measured on the FIQ-E in Manz et al.'s (2004) study are consistent with those measured on the FIQ-HS (school-based involvement, home-based involvement, home-school communication). A difference is that 40 items

Scales
Three scales were created using the 40 items based on the factor structure: homeschool communication, home-based activities, and school-based activities. The internal consistency of these scales was examined and found to be high to acceptable, with coefficients of 0.90, 0.88, and 0.76, respectively. Figure

Family characteristics and FIQ-HS
A one-way MANOVA was conducted to determine if any family characteristics were related to the three scales of the FIQ-HS. All 40 items were included in the scales, and division of items was done based on their placement on the factor analysis, for the 15 items that did not load onto any of the three factors researcher judgment was used to assign the item. The internal consistency of these scales was examined and found to be high to acceptable (α = .90 to 0.76).
Using Wilks Lambda statistic, there was a significant effect for the student's special education status on the three scales, Λ = 0.93. F(6, 912) = 5.60, p <.01. Parents who indicated that they did not know if their child was receiving special education services reported lower ratings of involvement on the home-school communication scale when compared to the parents who indicated their child was receiving special education services, F(2, 458) = 5.14, p <.05. Parents who indicated that they were unsure of their

Home-School Communication
Home-Based Activities School-Based Activities child's special education status also reported significantly lower home-based activities than parents who reported their child was either receiving or not receiving special education services, F(2, 458) = 4.29, p <.05.
Using Wilks Lambda statistic, there was also a significant effect for the student's school on the scales, Λ = 0.85. F(12, 1249) = 6.65, p <.01. One of the five schools had significantly higher parent ratings on home-school communication [F(4, 474) = 12.71, p <.01] and school-based activities [F(4, 474) = 6.10, p <.01]. Parent age, family ethnicity, and parent's relationship to the student were found not to be related to participants' ratings on the three scales.

Findings
The purpose of this study was to validate a high school version of the FIQ by establishing internal consistency and a factor structure consistent with the previous versions. The results of the Cronbach's alpha indicate high internal consistency of the overall questionnaire, and high to acceptable internal consistency for each of the three scales. Results of the exploratory factor analysis supported a three-factor structure consistent with the early childhood, elementary, New Zealand, and early childhood shortform versions of the FIQ.
Results from the confirmatory factor analysis did not confirm an identical factor structure of the FIQ-HS and the FIQ-E. This may be because of the significant differences between the current predominantly White rural school sample and the sample used in Manz et al.'s (2004) FIQ-E validation study which was predominantly African Americans from urban schools. Through the exploratory factor analysis it was discovered that a consistent factor structure with the FIQ-E was evident in the current data. However, fewer items loaded on the FIQ-HS than the FIQ-E, which likely caused the failed confirmatory factor analysis.
Results from the exploratory factor analysis indicated that 11 items on the FIQ-HS loaded onto the home-school communication factor, compared to 13 in Manz et al.'s (2004) study. Nine FIQ-HS items loaded on the home-based activities factor compared to 15 items in the FIQ-E study, and 4 FIQ-HS items loaded on the school-based activities factor compared to 12 in the FIQ-E study. The lower number of items that loaded onto factors in the current study may be attributed to differences in behavior between the elementary and high school family populations. Sample size is not believed to be an issue, as the current study had 517 parents and the Manz et al. study had 444 parents.
Considering the number of items on the questionnaire (40), the current sample size is considered appropriate for conducting a factor analysis.
There were 16 items that did not load onto any factor in the current study, compared to only 6 in the Manz et al. (2004) study. This is likely attributed to a difference in response patterns between the current sample and Mans et al.'s sample.
Given the significant differences between the two samples' ethnicity and geographic location, differences in response patterns are not surprising. Items that are the same between the FIQ-E and the FIQ-HS but did not have consistent results may not be meaningful or sensitive enough to the current sample.
Of the 16 items that did not load onto any factors only two items were new additions, "I teach my teenager how to perform home-living skills (e.g., laundry, dishes, car maintenance)," and "I ensure that my teenager has resources available to research post-secondary opportunities (e.g., colleges, careers)." This suggests that a majority of the 6 new items added to the FIQ-HS were appropriate and captured unique involvement behaviors that were not seen at the elementary level.
The three scales created based on the factor structure were found to have high to acceptable internal consistency. These scales represent three unique facets of family involvement. When participants' responses were examined, the home-based activities scale was found to yield higher involvement ratings than the home-school communication and school-based activities scale. This difference may be explained by the expectation that teenagers become more independent as they enter young adulthood.
So parents may be communicating less with school staff, shifting that responsibility to their teenager. Parents may be less involved in school-based activities because of the lack in-school opportunities made available at the high school level. However, as parents become less involved in activities at the school and communicating with school staff, they appear to continue to provide support to their teenager in the home setting.
A significant difference was found with parents who indicated that they did not know if their teenager was receiving special education services and parents who indicated that they did know if they teenager was receiving these services. It is important to note that although this difference is significant, only 9 participants in the sample indicated that they did not know their student's special education status. So this finding should be interpreted with caution.
Parents who indicated that their child was receiving special education services indicated higher ratings on the home-school communication scale than parents who indicated that they didn't know if their student was receiving services. This difference may be because of the necessity for communication and/or the federally mandated communication tactics schools are required to perform with parents of disabled students.
For example, parents of students with disabilities may need to contact school staff to obtain information that their student is not capable of relaying. Federal and state laws require school staff to be in regular communication with parents of disabled students to get their input on their child's programing, update them on their progress and changes to programing, and inform them of their rights. It is also likely that parents who indicated that they did not know if their child was receiving special education services are less likely to be in communication with their school to clarify such confusion.
Parents who indicated that their teenager was either receiving or not receiving special education services indicated higher ratings on the home-based activities scale than parent who did not know if their child was receiving services. This difference may be because parents who indicated that they were unsure of their teenager's special education status are less involved in their child's education and home life. For the parents with special needs students group, this difference may be because students with disabilities typically require more direct assistance to carry out life activities, so these parents may need to provide more direct assistance at home such as helping with homework, teaching them how to perform living skills, and providing learning materials.

Limitations
There were several limitations to the current study. Most importantly, this was a pilot study, so results should be interpreted with caution, as they are considered preliminary. Replication studies are needed to verify the current findings. Although the sample size (N = 517) was appropriate for the analyses conducted, a larger sample size would be beneficial. Replication studies should seek to have a minimum of 400 participants, to appropriately run the factor analysis on this 40-item questionnaire.
A significant limitation to the current study is the homogeneity of the sample. The majority of participants were Caucasian or White in ethnicity, from rural areas, and mothers of the students. Non-White participants accounted for only 3.9 percent of the total sample. This disproportionate representation in ethnicity is quite different from Manz et al.'s (2004) FIQ-E andFantuzzo et al.'s (2000) FIQ-EC validation samples in which the participant's were predominantly African American and living in urban areas.
Considering that the current results are being compared to the results found with these samples, it is fair to say that the factor structure found in the FIQ-EC, E, and HS versions appears to be consistent across White and African American ethnicities and urban and rural areas. However, more research with diverse samples is needed.
The majority of the responders to the survey were mothers (79.7%), which is consistent with Manz et al.'s (2004) sample (79%). This disproportionality of mothers and fathers should be examined in future research. Differences between the participant's relationship to the student and their responses on the FIQ-HS were not found. However, with a larger sample of fathers possible differences between these groups could be more appropriately examined.

Future Directions
Overall, more research is needed to evaluate the generalizability of the FIQ-HS and its internal consistency and factor structure across multiple samples. As a pilot study, the current research can be used for comparison against other samples using the FIQ-HS.
By continuing to develop this questionnaire it can eventually be utilized by schools to examine their family involvement practices on a regular basis. This is especially important in the high school setting where there is limited research on family involvement practices and few psychometrically sound assessments available for schools.
The FIQ-HS was found to assess three dimensions of family involvement.
However, it is possible that even more dimensions that are unique to the high school setting exist and could be captured by this instrument. This may be particularly true as we demonstrated that home-school communication and school-based activities decreased in the current FIQ-HS sample when compared to Manz et al.'s (2004) FIQ-E sample. As these dimensions decrease in high school, it is possible that other dimensions of family involvement are increasing. Future research should explore other possible dimensions such as community and civil participation, leisure and recreation, and functional academics (e.g., money management).
Multiple modifications, item eliminations, and item additions were made to adapt the FIQ-E to the high school setting. To increase the FIQ-HS factor loadings, the 15 items that did not load onto any factor could be reexamined for appropriateness to high school families or possible rewording. Adding new items to the FIQ-HS may also increase the number of items loading onto the factors. In the current study, only 6 new items were added to reflect post-secondary education and training, careers, and dailyliving skills. As this is a growing area of need in high school more items reflecting the various tasks and behaviors in these areas could be added.
The results indicated that one school in the current sample had significantly higher involvement levels in the home-school communication and school-based activities scales. This school was the only charter school included in the sample. These differences in involvement were examined on a small scale in the current study with the charter school's sample size of 29. While each school system has unique family involvement practices, general differences between traditional and alternative learning centers, such as charter schools, can be examined. If consistent results are found on a larger scale, specific behaviors or programs that are occurring in these alternative learning centers can be further examined to find practices that can be utilized in traditional schools to increase involvement.
In 2013, Fantuzzo and colleagues validated a short form of the FIQ-EC. They were able to reduce the original 42-item questionnaire to only 21 items and still demonstrate acceptable internal consistency and a consistent factor structure. The current 40-item FIQ-HS takes approximately 10-15 minutes to complete. If this questionnaire could be reduced, it would likely increase the participant response rate as well as ease data analysis. This may be particularly important for schools that may eventually administer this questionnaire to parents without the assistance of researchers.
The FIQ now has an early childhood, elementary, New Zealand elementary, early childhood short form, and high school versions validated. Future research should seek to validate a middle school version of the FIQ for grades 6 th through 8 th . This student family population is too unique from the elementary and high school population to include them in the FIQ-E or FIQ-HS versions. A middle school version of the FIQ can be developed by examining the items on the FIQ-E and the FIQ-HS for appropriateness to the middle school setting, and considering other areas unique to this population.

Appendix A Family Involvement Questionnaire-High School Version
Directions: For each item, please circle how often (Rarely, Sometimes, Often, or Always) you perform the activity. Please select only one response for each item.
Please Note: For the purpose of this questionnaire, the word teachers may include a number of school staff, such as guidance counselors, principals, school psychologists, or school social workers. Additionally, if you have other children that are in 8 th grade or lower, please respond to these items only considering your high school child(ren).

Rarely Sometimes Often Always
Appendix D

School Administrator Permission Form
On behalf of the School Psychology Program at Minnesota State University -Mankato, thank you for your consideration and interest in our research. We have received approval from the Institutional Review Board at Minnesota State University-Mankato and are seeking to move forward gathering family involvement data in your school district. At this time we are asking for your permission to proceed with research.
It is our plan to share these outcomes with the participating schools and possibly make recommendations on how each school site can improve their family involvement practices. This research is being conducted by Katlyn Grover, who will be directly supervised by Dr. Daniel Houlihan. We are asking for your participation in this research because we feel this would be of great use and benefit to your schools to use the information we obtain to improve family involvement practices in your high school. We ask that you read this form before agreeing to participate in the research.

Purpose
Past research has demonstrated that family involvement in education is an important contributor to children's school success. Children from families that have high rates of involvement typically earn higher grades, perform better on standardized assessments, earned more course credits, and are more likely to remain enrolled in school and graduate (Catsambis, 1988;Fan & Chen, 2001;Jeynes, 2005). Empirical investigations on family involvement in secondary schools are limited. However, there is a strong base of research investigating family involvement in early childhood and primary school settings. What we expect to see is that these factors of family involvement found in elementary settings are comparable to factors in secondary settings.
The purpose of this study is to validate the Family Involvement Questionnaire for use in high school settings. Responses to this questionnaire will also give us insights as to specific practices that may be manipulated or interventions that schools may put into place, which can increase family involvement.

Procedures
If you agree to participate in this research by signing this consent form, we ask that you release contact information for your 9 th through 12 th grade parents to the primary investigator, or disperse mailing packets and/or survey links to the appropriate parents directly. If you choose to have us disperse the research materials, then the contact information we require would be either parent email addresses or mailing addresses. We will then contact the parents for their voluntary informed consent to participate in this study. Parents who choose to participate in the study will complete two short questionnaires, the Family Involvement Questionnaire and a demographic questionnaire.
The Family Involvement Questionnaire will ask parents about their involvement with their child's school and academic well-being. The demographic questionnaire will ask basic questions about the parent, their child(ren), and their family's make-up. These two questionnaires should take approximately 20 minutes to complete.
If you choose to have parents contacted by postal mail, then parents who agree to participate will be asked to return their completed Family Involvement Questionnaire and demographic questionnaire to us via an envelope provided to them.
If you choose to have parents contacted through email, then parents who agree to participate will be directed to an online survey system (through Qualtrics). Parents will initially be directed to a consent form, which they will need to agree to before they may complete the Family Involvement Questionnaire and demographic questionnaire. Once they have completed both questionnaires online, their responses will be sent to a secured online database were we can retrieve them.

Risks and Benefits
There is little risk involved with the study. However, some parents may experience feelings of embarrassment for the answers that they may choose. This risk should be mitigated by the fact that responses will be anonymous.
Although, responses will be anonymous, whenever one works with online technology there is always the risk of compromising privacy, confidentiality, and/or anonymity. If you or a participating parent would like more information about the specific privacy and anonymity risks posed by online surveys, please contact the Minnesota State University, Mankato Information and Technology Services Help Desk (507-389-6654) and ask to speak to the Information Security Manager.
There are no direct benefits for the parents participating in this research, but the results of this study may help guide your school to improve family involvement practices.

Confidentiality
All records of this research will be kept private. All assessment data and parent contact information will remain confidential. In any sort of report, we will not include any information that will make it possible to identify a parent or school. Parent informed consent forms are anonymous, meaning we will not be collecting any individually identifiable information from participating parents such as names or birthdays.
School privacy will be maintained by using a false name for the school, and all other individuals that may be identifiable in this research. Additionally, these false names will be used in all communications and all research published from this data.
All forms for this research will be kept in a locked file cabinet in a research storage room at Minnesota State University-Mankato. Only the researchers have access to the records. These records will be kept for 3 years, and then they will be destroyed.

Voluntary nature of the research project
Your decision whether or not to participate in this research will not affect your current or future relations with the School Psychology Program at Minnesota State University, Mankato, or the researchers. Even if you sign this permission form, you are free to stop participation at any time.
You may print a copy of this page for your future reference.
Thank you so much for your consideration!

PARTICIPATION
( YES ) I agree to participate in this study.
If you do not want to participate in this study you may close this web browser.

March 2, 2015
On behalf of Minnesota State University-Mankato and the School Psychology Doctoral Program we want to thank you for your participation in our family involvement study. This winter we had five high schools in southern Minnesota participate in our study, and over five hundred parents complete our online survey. This is your school district's individual feedback report. You will find data analyzed from your responding parents as well as data from our entire sample.
 On page 5 you will find demographic information of your responding parents.
 On page 6 you will find the average parent responses for the three scales; homeschool communication, home-based involvement activities, and school-based involvement activities. This area provides an overall view on how your parents' rated their level of involvement in each area. A comparison to our entire sample is also provided.
 On pages 7 to 10 you will find your parent responses for each of the 40 items on the questionnaire. You can examine each item by looking at the mean response and the distribution of responses across the four choices (rarely, sometimes, often or always).
 On page 11 you will find a strength and weakness report. This section will provide specific behaviors your parents' rated as often or always performing, and behaviors they rated as sometimes or rarely performing.
 On pages 12 to 15 you will find data obtained from our entire sample. While each school site has unique family involvement practices, we understand that you may be interested in comparing your school's results to others in our sample. Individual item responses for all participating parents' are provided.