PROTOCOL: Interventions Intended to Increase School Attendance in Elementary and Secondary School Students

Definitions

There is a substantial body of literature related to school absenteeism; however, there is a lack of consensus and “considerable disparity about fundamental concepts of definition and meaning, assessment, and treatment” (Kearney, 2003, p. 57). Terminology utilized in this body of literature includes truancy, school absenteeism, school refusal behavior, school phobia, anxious school refusal, problematic absenteeism, and school non-attendance. There is no universally agreed upon definition for any of the terms used in the literature. Many terms are used interchangeably, are used differently by different authors and have evolved over the years. However, I will attempt to provide an overview of the terms and definitions frequently utilized in the literature to provide some understanding of how the terminology is utilized and in what contexts.

Truancy

Truancy is applied as an overall descriptive term for students who are absent from school for one reason or another, as well as used as a legal term referring to absences that are illegal as defined by statute. Broadwin (1932) broadly defined truancy as “absence from school without proper leave” (p. 253), citing various reasons why one might be absent from school. Reid (1999) more specifically defined truancy as “miss[ing] school illegally, with or without the consent of their parent” (p. xi). Some authors distinguish truancy from other forms of absenteeism as an absence which is unexcused, is done without the knowledge of the parent and is not due to anxiety or fear (Kahn, Nursten, & Carroll, 1981; Lauchlan, 2003). Truancy has also been used to refer to students whose parents keep their child home to work, take care of siblings, etc. (Kahn et al., 1981). Kearney (2008b) defined truancy as “unexcused, illegal, surreptitious absences, non-anxiety-based absenteeism, absenteeism linked to lack of parental knowledge about the behavior, absenteeism linked to delinquency or academic problems, or absenteeism linked to social conditions such as homelessness or poverty” (p. 452).

In addition to the various ways in which truancy is used in the scholarly literature, the term truancy also carries local meaning (Reid, 1999). Different states have different compulsory education laws, thus making the definition of the term dependent upon state statues. School districts, and even different schools within the same school district, have different definitions of truancy and different standards for when they consider a child ‘truant’ (Garcia-Gracia, 2008). This makes it difficult, perhaps impossible, to find consensus regarding the meaning of the term truancy or to compare rates of truancy from one state or school district to another.

School Refusal Behavior

The conceptualization and definition of school refusal behavior has evolved over the years and has been employed differently by various authors/researchers. Kahn et al. (1981) defined school refusal as “cases where there is a psychosocial component” (p. 3). King and Bernstein (2001) define school refusal as “difficulty attending school associated with emotional distress, especially anxiety and depression” (p. 197). Kearney and Bates (2005) define school refusal behavior as “any refusal to attend school for an entire day by a child” (p. 207) and include youth who “miss long periods of school time; miss sporadic periods of school time, skip classes, or arrive tardy to school; or attend school with great dread and somatic complaints that precipitate pleas for future non-attendance” (p. 207). Kearney (2007) defined school refusal behavior as an “umbrella term that covers many hypothesized subtypes of youths with problematic absenteeism, including truancy, school phobia and anxiety-based school refusal” (p. 53). King & Bernstein (2001) defined school refusal as “difficulty attending school associated with emotional distress, especially anxiety and depression” (p. 197).

School refusal behavior is often distinguished from truancy by 1) an absence of antisocial behavior/characteristics; 2) parental awareness of the problem and knowledge of the absence from school; and 3) presence of emotional distress, separation anxiety, anxiety and/or depression (Elliot, 1999; King, Tonge, Heyne, Pritchard, Rollings, Young, et al., 1998; Heyne, King, Tonge, & Cooper, 2001). There is considerable debate in the literature as to whether school refusal behavior should be more broadly used to encompass truancy as Kearney (2007) suggests or whether school refusal behavior should be distinguished from truancy as a different type of school attendance problem as Heyne et al. (2001) recommends.

School Phobia

The term school phobia is applied to describe students who are not attending school due to the fear of going to school and who meet DSM criteria for specific phobia (Fremont, 2003). Although the definition of school phobia is probably the most concrete of all the terminologies used in this body of literature, school phobia is often used interchangeably with school refusal behavior or gets subsumed under this more broad term (King & Bernstein, 2001). Kearney (2008) noted that the prevalence of students being phobic of school is rare and thus the term has been deemphasized in the literature.

School Absenteeism and School Non-Attendance

School absenteeism and school non-attendance are broad terms used interchangeably to describe an occasion when a student misses school, regardless of reason. Kearney (2008) defines absenteeism as “excusable or inexcusable absences from elementary or secondary (middle/high) school” (p. 452). School absenteeism and school non-attendance are more neutral terms than truancy, school refusal behavior and school phobia, as the former do not carry the emotive connotations associated with the latter (Reid & Kendall, 1982). Truancy, school refusal behavior and school phobia are all based on a pathological model, with non-attenders either being viewed as ‘mad’ or ‘bad’ (Carlen, Gleason, & Wardhaugh, 1992). Authors who utilize the terms school absenteeism and school non-attendance argue that these terms provide a non-pathological conceptualization of the problem and advocate the use of these terms over truancy and school refusal (Lauchlan, 2003; Lyon & Cotler, 2007; Pellegrini, 2007).

Discussion of Terminology

This brief overview of the terminology utilized in this body of literature highlights the lack of shared definition and conceptualization of the problem of school non-attendance. Although the literature often differentiates school refusal from truancy based on the reasons students are not attending school and whether or not the absence was known by the parents, some authors have argued that the distinction between the two terms and whether or not the absence was excused is unnecessary, counterproductive and logistically difficult (Lauchlan, 2003; Kearney, 2008; Lyon & Cotler, 2007; Pellegrini, 2007). In addition, the differences between students who are classified as truant and those classified as school refusers are not clear cut. There is considerable diagnostic heterogeneity in both groups and substantial overlap in symptoms (Egger, Costello, & Angold, 2003; Kearny, 2008). There is also evidence that some students can exhibit both truant behavior as well as school refusal behavior either concurrently or sequentially (Berg, Butler, Franklin, Hayes, Lucas, & Sims, 1993; Bools, Foster, Brown, & Berg, 1990).

The utility of classifying students as excused or unexcused has come under debate. Some argue that the outcomes for students, schools and communities are the same regardless of the reasons for students missing school or if the absences were known by the parent (National Center for School Engagement, 2007). Eaton, Brener, & Kann (2008) found that absentee students, regardless of whether they had permission or not to miss school, are more likely to engage in risk behaviors than those with no absences. Malcolm et al., (2003) also argue that distinguishing between authorized an unauthorized absences in unhelpful. Schools apply the terms differently and accept a range of evidence for authorizations, thus making the distinction invalid, or at least inaccurate. Also, parents, or clever students posing as a parent, may provide an excuse for an absence after the fact, thus validating an absence as excused when it really was not. Malcom et al. (2003) argued that classifying absences in this way only masks the scale of the problem, thus reducing the imperative to seek solutions to the problem.

Lauchlan (2003) and others argue that the problem of school non-attendance is heterogeneous and we should not be bogged down in making invalid and unnecessary distinctions when addressing the problem. Because of the conflicting, confusing and changing constructs and definitions used for school refusal behavior, truancy and other terminology, the categorical distinctions perpetuated in the literature have not necessarily been useful when responding to the problem (Kearney, 2003; Lauchlan, 2003).

For the purposes of this systematic review, the term school absenteeism will be utilized throughout the review. School absenteeism will encompass the subcategories of research related to truancy, school refusal or other research examining the problem of students missing school for whatever reason, except when citing a specific study or body of literature that utilizes a different term/definition, to allow for a broad range of studies to be included.

Prevalence of Absenteeism

Accurately reporting the prevalence of absenteeism is challenging because of inconsistent reporting requirements and the use of multiple definitions and constructs of truancy, school refusal and school absenteeism (Kearney, 2003; Lyon & Cotler, 2007; Pellegrini, 2007). It is estimated that hundreds of thousands of youth are not attending school on a regular basis, many without an excuse (Baker, Sigmon, & Nugent, 2001). Several large inner-city schools systems report thousands of unexcused absences each day with some reporting absentee rates as high as 30% (Garry, 1996). According to the National Center for Education Statistics (2006), 19% of students in 4^th grade and 20% of students in 8^th grade reported missing three or more days of school in the preceding month. The study also noted that the patterns of absenteeism have remained relatively stable between 1994 and 2005. According to recent statistics available from the U.S. Department of Justice, the number of truancy cases petitioned and handled in juvenile courts increased 69% between 1995 and 2004, and accounted for the largest proportion (35%) of status offense petitions handled by the juvenile courts (Stahl, 2008).

Costs of Absenteeism

The problem of school absenteeism has several implications for the youth who do not attend school regularly as well has his/her family, school and community. The negative outcomes for the truant/absentee youth include delinquency, poor school performance, school expulsion and dropout, substance use and other risky and problematic behaviors (National Center for School Engagement, 2007; Petrides, Chamorro-Premuzic, Frederickson & Furnham, 2005; Reid, 1999). The economic implications for students are also significant. Students who are chronically absent are more likely to perform poorly in school and more likely to drop out, which negatively impacts their earning potential over their lifetime (Attwood & Croll, 2006; Garry, 1996). The implications for schools whose students are not attending at a high rate include loss of funds and failure to meet performance requirements (Goldstein, Little, & Akin-Little, 2003). Significant costs to communities associated with truancy/absenteeism include higher rates of criminal activity, citizens not productively contributing to the community and higher government spending for social services (Baker et al., 2001).

Overview of Absenteeism Research

Due to the seriousness and prevalence of the problem of absenteeism, researchers from several different fields, including social work, education, psychology, nursing and criminal justice, have attempted to understand and address the problem. A large body of literature has been accumulating over the past several decades related to the causes, correlates, outcomes and interventions with this population.

The causes of school absenteeism have been given extensive attention in the empirical research in the field. Research indicates a number of factors that have demonstrated some causal or correlational relationship to school absenteeism. These include individual, family, school, and community/contextual factors.

Individual risk factors associated with absentee youth include lower academic self-concepts, lower self esteem, less competent social relations, phobia, anxiety, personality traits, race/ethnicity and learning disabilities (Corville-Smith, Ryan, Adams & Dalicandro, 1998; Lounsbury, Steel, Loveland & Gibson, 2004; Malcolm, Wilson, Davidson & Kirk, 2003; Romero & Lee, 2008; Sheppard, 2005; Southwell, 2006).

School factors identified as causal or correlational to absenteeism include school culture, curriculum, poor teaching, negative school environment, interpersonal conflict or poor relationships with teachers, dissatisfaction with school, school disciplinary practices, and threats to physical safety such as bullying (Corville-Smith et al., 1998; Enomoto, 1994; Malcolm et al., 2003; Reid & Kendall, 1982).

Family factors, such as family conflict, poor/unhealthy family relationships, parental attitudes and values toward education, lack of cohesion, inconsistent and ineffective discipline, sanctioning/colluding of school absences by parents, parent-child interactions, parental involvement in school, family poverty and family structure have been implicated as causal/correlational factors in absenteeism research (Corville-Smith et al., 1998; Malcolm et al., 2003; McNeal, 1999; Romero & Lee, 2008).

Community/contextual factors have also been found to have effects on school absenteeism. These factors include race/ethnicity, socio-economic status, employment and other opportunities in the community, neighborhood characteristics and level of organization, levels of social support, community norms, and community violence (Bowen, Bowen, & Ware, 2002; Lyon & Cotler, 2007; MacDonald & Marsh, 2007).

School absenteeism is increasingly being recognized as a complex and heterogeneous problem that can be influenced by a number of factors (Kearney, 2008; Kim & Streeter, 2006; Lauchlan, 2003). Researchers and practitioners have developed various strategies targeting a number of risk factors that have been associated with absenteeism resulting in diverse intervention strategies being implemented in various settings.

Interventions to Increase Student Attendance

The number of interventions designed to increase student attendance have been growing substantially. In the United States, several federal and community initiatives have been established to reduce absenteeism. The Office of Juvenile Justice and Delinquency Prevention (OJJDP) established a database of programs that have shown some effectiveness in reducing truancy. Included in this database, the Model Programs Guide, are 16 model programs. The National Center for School Engagement lists 171 truancy programs registered in their database, of which 69 have had external evaluations and 30 have final evaluations completed.

In addition to national initiatives to improve attendance, several national initiatives have also been implemented to reduce high school dropout. Although this review is concerned with absenteeism rather than the problem of school dropout, absenteeism is strongly associated with, and has been identified as a significant risk factor for, school dropout (Baker, 2001; Garry, 1996). As a result, many strategies utilized in dropout prevention programs focus on increasing student attendance, thus there is some overlap between absenteeism and dropout interventions. The National Dropout Prevention Center, for example, lists 60 model programs for truancy reduction in their database. This review will focus on interventions intended to increase student attendance, thus will likely include some studies of interventions that are identified as “dropout prevention” programs. Not all dropout prevention programs have an identified goal of increasing student attendance and do not measure attendance, thus many studies of dropout prevention programs will not meet criteria for this study.

Because school absenteeism is a recognized problem among various disciplines, including education, psychology, social work, nursing, criminal justice, sociology and others, the conceptualizations of the problem as well as the approaches used to intervene with school absenteeism are diverse. Interventions targeting school attendance fall into several different categories, target a variety of different risk factors and levels, are implemented in different settings and are delivered through a variety of modalities. Interventions generally target individual risk factors, such as anxiety/phobia, low self-esteem, social skills and medical conditions; family factors, such as communication and parental support, discipline/contingency management, parental involvement and communication with the school; and school factors, such as school climate, attendance policies, relationships between teachers and students and bullying. Several interventions target multiple risk factors across all three levels.

In addition to the variety of risk factors targeted, interventions also differ in terms of the settings in which the interventions are implemented. Interventions have been implemented in clinical and community agency settings, schools, courts and police agencies. Interventions may be conducted as part of a collaborative effort between community agencies, schools, courts and/or police agencies or by a single entity.

Depending on the risk factor(s), the level being targeted and the setting(s) in which the intervention is being carried out, programs intended to increase student attendance are delivered in a variety of modalities. These include, but are not limited to individual therapy, parent training, family therapy, group therapy, monitoring/supervision, case management, incentives/rewards, fines/sanctions, prosecution, social service referrals, tutoring, teacher training/development, and school improvement strategies.

Despite the widespread attention to this problem and the increase in interventions designed to reduce absenteeism, the issue remains a significant problem. The lack of consensus about definitions and conceptualizations of the problem as well as intervention strategies for youths with problematic absenteeism has contributed to the disconnection between sets of professionals studying absenteeism. While examining a problem from various perspectives can be productive, the study of absenteeism has remained disparate. Several authors in various fields studying the problems of student non-attendance have concluded that the problem of non-attendance is heterogeneous and lies along a continuum, thus maintaining a distinction between truancy and school refusal is unnecessary and can be counterproductive (Kearney, 2008; Lauchlan, 2003; Lyon & Cotler, 2007). They have called for a more inclusive and integrated conceptualization of absenteeism and a need to include all students exhibiting problems with absenteeism in research, assessment and treatment. Thus, this review will utilize a broad conceptualization of absenteeism.

Prior Reviews of Programs Targeting Truancy, School Refusal, and/or Absenteeism

A search for previous reviews and meta-analysis of interventions related to the problems of school absenteeism, school refusal, school attendance, school non-attendance and truancy was undertaken. Six databases (ERIC, PsychInfo, Academic Search Premier, Dissertation Abstracts, Criminal Justice Periodicals and Pegasus (Loyola's book search), were searched and twenty-two reviews were identified. A summary of the findings of the search will be discussed below. Twenty-two reviews were identified (see Appendix A). There were no meta-analysis identified.

Of the twenty-two reviews identified, 17 were traditional narrative reviews of the literature. Because these 17 reviews were very similar in their nature and content, the individual reviews will not be discussed in great detail. These literature reviews focused on literature regarding causes, correlates, diagnostic features, etc. as well as highlighted various treatment modalities, citing published intervention studies to provide evidence of effectiveness of the treatments included in the report. Much of the discussion of intervention in these reviews covered a range of programs and settings, providing discussion and descriptions of different types of interventions available. These reviews cited relatively few studies of interventions.

In addition, none of these 17 reviews were systematic. They did not specify their search strategy or inclusion/exclusion criteria and they included only published studies. The outcome studies that were cited in the reviews used various methodologies including case studies, open clinical trials, randomized and non-randomized studies (see Appendix B). The reported findings primarily favored the intervention discussed. Many of these reviews also focused on the same literature base, yielding considerable repetition. This is especially the case as it relates to the reviews related to the subgroup of school refusal, which tends to cite the same studies on cognitive behavioral treatment and pharmacology.

A narrative review of strategies to encourage attendance conducted by Railsback (2004) was more comprehensive and inclusive than the above traditional narrative reviews. Railsback “surveyed the last decade of research that discusses strategies or experiments to increase student attendance” (p. iii). The author did not describe her search strategy, but did include unpublished studies. The author's stated intent was to include only “scientifically based” research, but broadened her inclusion criteria to include a range of research designs as well as surveys and expert opinions because “it was quickly determined that little research of that kind [scientifically based] exists” (p. iii). The reviewer concluded that “we found no research that definitively answers the question: Do some strategies [to encourage attendance] work better than others?” (Railsback, 2004, p. 11). The author then summarized the literature and strategies found during the search and provided guidelines to readers interested in implementing policies and programs to increase student attendance.

Three of the 22 reviews used narrative synthesis approaches and were narrow in focus. Two were of cognitive-behavioral interventions dealing with the treatment of school refusal (King, Tonge, Heyne & Ollendick, 2000; King, Heyne & Ollendick, 2005) and one was a review of welfare-school attendance programs (Cambpell & Wright, 2005). The 2000 review by King and colleagues included eight studies. The authors’ search strategy was not specified; however, all studies in the review were published. Their inclusion/exclusion criteria entailed including only cognitive-behavioral interventions and all methods. The report concluded that “At first glance, our review of research suggests empirical support for cognitive-behavioral therapy in the treatment of school refusal …” (p. 501). “However, since very few controlled studies have been reported at this stage in treatment research, it would be premature to extol the clinical virtues of cognitive-behavior therapy” (King et al., 2000, p. 506).

The 2005 review by King and colleagues focused on a broader topic of anxiety and phobic disorders, but did review seven studies on school refusal behavior. They limited the studies included to those studying behavioral or cognitive-behavioral treatments. The search strategy for this review included searching literature in peer-reviewed journals from 1980, but did not specify the year they used for the cutoff. The authors did not specify which journals or databases they searched, but provided some examples of specific journals they included. Of the seven studies included in this review, five of them were used in the previous review. One of the additional studies was a follow-up study of a randomized control trial included in the previous review and the other was a randomized trial with a comparison group which received an alternative treatment (which showed no significant difference between CBT and the alternative treatment). Although the authors used substantially the same studies in both reviews, they came to a different conclusion. The authors concluded that “overall, school refusal has responded to CBT programs as demonstrated in a number of controlled studies, with general maintenance of gains” (King et al., 2005, p. 249).

The final review identified was a narrative listing of 23 model, promising and emerging truancy programs published by the National Dropout Prevention Council (Reimer & Dimock, 2005). The programs chosen for this publication were based on the author's familiarity with the program and if the program “demonstrated success and practicality of implementation in a variety of environmental realities and programmatic contexts” (Reimer & Dimock, 2005, p. 7). The authors did not specify the criteria they used to establish a program as a “success” and did not provide outcome data or method of evaluation in their narrative descriptions of the programs.

Reviews and meta-analyses have also been conducted of interventions to target other related school problems, such as problem behaviors, school performance, and anxiety and phobic disorders, with school attendance being one of the measures used (King, Heyne, Ollendick, 2005; Little & Harris, 2003; Maughan, 2003; Mattison, 2000; Wilson, Gottfredson, and Najaka, 2001). However, not all studies included in these reviews and meta-analyses measured attendance as attendance was not the primary problem being reviewed. Thus there is limited information in these reviews related to interventions intended to increase attendance per se.

In addition to published reviews of interventions targeting school attendance, lists of “model” truancy reduction programs have been developed by the OJJDP, the National Center for School Engagement (NCSE) and the National Dropout Prevention Center (NDPC). Only the OJJDP database of model programs specifies criteria for inclusion of programs in the database. They rate programs as exemplary, effective or promising dependent upon the rating criteria. Programs classified as exemplary must have demonstrated effectiveness using an experimental design; programs classified as effective must have demonstrated effectiveness with a quasi-experimental design; and programs classified as promising have demonstrated effectiveness using limited evaluation designs such as single group pre-post test designs. The rating and classification system is also based on four dimensions of program effectiveness: 1) conceptual framework of the program; 2) program fidelity; 3) evaluation design; 4) empirical evidence demonstrating the prevention of problem behavior; the reduction of risk factors related to the problem behavior; or the enhancement of protective factors related to problem behavior (OJJDP Model Programs Guide). The National Center for School Engagement's database is a self-registry of programs, requiring no minimum criteria to be met in order to be registered in the database. Thus programs that are ineffective could be listed amongst those that have demonstrated effectiveness. The National Dropout Prevention Center's list of model programs does not specify criteria by which the programs have been determined to be “model” programs.

Although having lists of programs in various databases may be helpful at some level, merely listing programs with varying levels of evaluation and evidence of effectiveness can be misleading to those who are looking for programs to implement. A review and synthesis of these outcomes of interventions, using what is likely unpublished evaluations of the programs, is needed to summarize the extant research in this area, estimate the magnitude of the program impacts (effect size) and establish the evidence base for programs being disseminated through these guides and registries.

The knowledge being gained about interventions to increase student attendance is growing substantially. From the literature reviews and lists of “model” programs, there seems to be a number of diverse programs that have been evaluated, both published and unpublished, providing a substantial body of research available for assessing the efficacy of interventions to increase student attendance. Unfortunately this knowledge is disparate, confusing, and much is possibly unpublished, making it difficult for policy makers and practitioners to use evidence of effectiveness to guide policy and practice.

Summary

To date, we have not been able to locate a meta-analysis or systematic review of interventions intended to increase school attendance in primary or secondary school students. It is important to synthesize the intervention research to provide a comprehensive picture of interventions that are being utilized, to identify interventions that are effective and identify areas in which more research needs to be conducted to better inform practice and policy. This review will fill this gap in the literature with the ultimate goal of providing evidence-based guidelines to help guide policy makers and practitioners in helping students attend school regularly.

Benefits of the Proposed Review

The proposed systematic review will improve upon prior work in several ways. First, this review will apply a systematic and transparent process for searching, retrieving and coding studies. Utilizing a systematic method to conduct the review of outcome research limits bias and reduces chance effects, leading to more reliable results (Coooper, 1998). The application of explicit and transparent description of the review process also allows for the review to be replicated and expanded to include new studies or criteria.

Second, this review will include evaluations of interventions operating in a broader set of geographical contexts than previous reviews, including programs across the United States as well as other countries with similar educational systems. This will allow us to potentially identify studies that have been missed in prior reviews.

Third, this review will include evaluations of interventions targeting student attendance, rather than being bounded by specific conceptualizations of truancy or school refusal behavior which have defined prior reviews. Researchers in this field have often made a distinction between “truancy” and “school refusal”, thus reviews are often specific to either truancy or school refusal. Several authors in various fields studying the problems of student non-attendance have concluded that the problem of non-attendance is heterogeneous and lies along a continuum, thus maintaining a distinction between truancy and school refusal is unnecessary and can be counterproductive (Kearney, 2008; Lauchlan, 2003; Lyon & Cotler, 2007). They have called for a more inclusive and integrated conceptualization of absenteeism and a need to include all students exhibiting problems with absenteeism in research, assessment and treatment. Thus, this review will include studies of interventions targeting student attendance.

Fourth, we will evaluate whether the research base is an adequate representation of programs currently in operation. Although we will not be systematically assessing all programs in operation, we will rely on summary reports by government and non-government entities to inventory strategies aimed at increasing student attendance. We will determine the extent to which there is credible evidence of the impacts of these particular strategies by comparing programs in operation and recommended intervention strategies with the studies included in this review. We will also explore differences in outcomes among clusters of programs defined by seemingly important programmatic features and assess the appropriateness of combining effect sizes for different types of programs.

And lastly, prior reviews have been limited to a narrative approach, presenting a description of programs or using a vote-counting method to categorize outcomes of programs as significantly positive, significantly negative or no significance. Conclusions regarding effective interventions can then made based on the number of studies that were found to demonstrate significant positive results. The vote-counting method, however, disregards sample size, thus leading to erroneous conclusions (Glass, McGaw & Smith 1981). Also, the vote-counting method relies on statistical significance and does not take into account measures of the strength of the study findings, thus also leading to misleading conclusions (Glass, McGaw & Smith, 1981). Meta-analysis, on the other hand, represents key findings in terms of effect size rather than statistical significance. Thus, meta-analysis provides information about the strength and importance of a relationship, the magnitude of the effects of the interventions and the characteristics of effective interventions.

4.1 Criteria for Inclusion and Exclusion of Studies in the Review

The following criteria will be used to determine whether a study will be included in the review for purposes of estimating program effects:

Types of studies: Randomized Controlled Trials, Quasi-Experimental Designs and single group pre-post test designs will be included in the review. Single group pre-post designs will be included in the review as many of the programs being evaluated through the OJJDP initiatives and other unpublished studies/evaluations are likely utilizing this methodology. Also, prior reviews indicate a dearth of randomized control trials and quasi-experimental designs. The studies utilizing a single group pre-post design will be analyzed separately from RCTs and Quasi-Experimental Designs.

Case Study designs will also be included in the review and coded for quality; however, case study designs will not be included in the analysis. Case studies appear to be utilized often in school refusal and truancy research. The purpose of searching for and including case study designs is to provide a fuller picture of strategies that are being utilized in the field and to determine if the research base adequately represents the range of programs currently in operation. Finding a number of case studies that demonstrate promising results with certain interventions could provide a basis for recommending further research.

4.2 Search Strategy for Identification of Relevant Studies

We propose to include all studies that meet the inclusion criteria outlined above. An attempt will be made to identify and retrieve both published and unpublished studies. Several sources will be used to identify eligible studies, including: 1)

Electronic databases;

Databases

Academic Search Premier

Within each database, we will search using combinations of keywords grouped into four main categories: 1)

Outcome: Attendance OR Absen*

AND

We will adapt the search strategy for different databases, consulting database thesauri when they are available to ensure that the appropriate synonyms have been included.

Internet searches

Websites of all relevant government, research centers, foundations and professional associations will be searched for published and unpublished studies. Some relevant websites include coloradofoundation.org, hfrp.org, truancyprevention.org, drgonline.com, Colorado.edu/cspv/blueprints/, schoolengagement.org, dropoutprevention.org, ies.ed.gov/ncee/wwc/.

Personal contacts

The reviewers will contact researchers and experts in the area of truancy/school absenteeism/school refusal in an effort to uncover additional published or unpublished studies relevant to the review.

Reference lists

Reference lists of prior reviews and related meta-analyses will be reviewed for relevant studies. In addition, the references of the retrieved primary studies will be examined for potential studies relevant for the review.

4.3 Conducting and Documenting the Search and Selection Process

A comprehensive search log will be maintained to keep track of all searches including 1)search engines used; 2) Database or main source searched; 3) key words used; 4) time period searched; 5) number of hits; and 6) time spent searching. A list of study titles, citations and inclusion decisions will be documented and managed using Excel software. All titles and abstracts found through the search procedures will be reviewed and screened by one of the reviewers. The abstracts will be reviewed primarily for relevance, with final eligibility screening based on the entire article. For example, studies that would be deemed as inappropriate at the abstract screening stage would be those that do not involve the target population (e.g. college students or adults) or are descriptive or theoretical in nature where no intervention is being evaluated. All abstracts deemed potentially appropriate will be retrieved in full text. Also, if there is any question as to the appropriateness of the study at this stage, the full text will be obtained.

Once the full text of the studies are retrieved, each study will be reviewed and the basic information needed to determine whether the study meets the overall inclusion criteria as described above will be coded. For those studies that meet the inclusion criteria, we will then code the information required to determine quality of study implementation and the information required for the analysis of program impacts for the total sample and for key subgroups and program types. All studies meeting the inclusion criteria will be summarized in a table.

Studies will be retrieved primarily through the Loyola University Chicago library system, InterLibrary Loan, University Microforms and electronic databases and websites listed above. When possible, PDF files of all articles will be saved in an electronic folder. If PDF or electronic versions are not available, hard copies will be kept on file.

4.4 Description of Methods Used in Primary Research

The most common methods used in evaluating interventions to reduce absenteeism/increase attendance includes quasi-experimental designs and within group pre-post designs. The quasi-experimental studies compare the attendance of a sample of students who received an intervention to a matched sample of those who did not receive the intervention. The single group pre-post studies compare attendance rates prior to the intervention and following the intervention. The most common outcome is attendance as measured by number or percentage of days absent from/present in school. Some studies measure attendance through self report asking the student how often they were absent from school during a certain time period. In these cases, attendance is often measured using a Likert scale (0 days absent, 1-3 days, 4-6 days, etc.). It is very uncommon for studies of interventions aimed at impacting attendance to use dichotomous outcomes.

An example of a quasi-experimental study using a matched comparison group to evaluate the effectiveness of an intervention to increase attendance was employed by Sheldon (2007). Sheldon examined whether implementing the National Network of Partnership Schools (NNPS) program affected student attendance. Sheldon compared sixty-nine NNPS elementary schools with a matched sample of 69 non-NNPS schools. The schools were matched on a) student performance on mathematics and reading achievement tests; b) student enrollment and c) average daily attendance. He also compared the schools on racial composition, percentage of students receiving free/reduced lunch, Title 1 status and pupil-teacher ratio. A comparison of the NNPS schools and non NNPS schools suggested that the two groups were similar in many ways, but did have differences in 1) NNPS schools tended to have higher percentage of students receiving free/reduced lunch; 2) NNPS schools tended to serve more disadvantaged families; 3) non-NNPS schools spent a great percentage of their total budget on pupil support; 4) NNPS schools tended to have slightly lower rates of daily student attendance. The author included the following variables as covariates: pupil funding, student race, family income and pupil-teacher ratio. The author conducted multiple-regression analyses, using NNPS status as a treatment-dummy variable, to examine the effect of schools implementing NNPS on student attendance in 2001 and change in rates of student attendance from 2000 to 2001. After controlling for several variables described above, the author found that schools implementing the NNPS partnership program had higher levels of student attendance than did the matched sample of non-NNPS schools. Calculation of effect sizes suggested that school-wide partnerships partnership programs may have a small-to-moderate effect on student attendance.

4.5 Criteria for Determination of Independent Findings

We are primarily interested in one outcome measure, attendance, which is usually measured in one way within a study (e.g. school records or student report), thus the issue of data independence should not be a factor in the analysis of the data. If a study provides data on additional outcome measures, we will extract and code the effect sizes for each measure and conduct separate meta-analyses.

In cases where there are multiple points of follow-up for a given outcome measure, we will record all points of follow up and conduct a separate analysis for effect sizes at similar points of follow-up. If there are an adequate number of studies with longitudinal follow-up, we will examine changes in effect sizes over time. We will code for retention at all data points and assess the level of retention in a moderator analysis.

In cases where we encounter studies with multiple outcomes for dependent or overlapping samples (e.g., multiple treatments compared against one control group), we will code all of the effect sizes but only include one treatment/control comparison in the meta-analysis. We will select the primary, or most relevant, treatment being tested to include in instances where both treatments are within the subgroup pooled.

Pilot programs and their replications will be treated as independent study samples. Replication status will be coded and potentially used in the analysis as a descriptive moderator.

4.6 Details of Study Coding

All studies meeting the initial criteria will be coded using an instrument which will specify the information to be extracted from each eligible study as described by Lipsey & Wilson (2001). The coding instrument will include items related to bibliographic information and source descriptors; methods and procedures; context, nature and implementation of the intervention; sample characteristics; and outcome data needed to calculate effect sizes (see Appendix C).

To ensure reliability of coding procedures and decisions, 100% of included studies will be independently coded by two individuals. If inter-rater differences occur, these will be discussed in order to refine coding schemes and resolve any discrepancies. A third author will be consulted and any remaining discrepancies resolved by consensus with a third author.

4.7 Statistical Procedures and Conventions