This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.
The goal of this study is to determine the effectiveness of discipline programs and methods used in reducing truancy in an alternative middle school. To this end, a mixed-methods research study is proposed. Data gathered from three sources will be utilized to include a review of school records, a survey questionnaire, and a face-to-face interview. This study will benefit from both qualitative and quantitative research methods which will assist the researcher in answering the study's research problems:
What are the current behavioral and situational factors that are predictive of truancy in middle school students of alternative schools?
How effective is the discipline program within alternative schools for targeting truancy in middle school students with respect to their behavioral and environmental needs?
How can the discipline methods be manipulated to target these behavioral and situational trends?
Section 2 describes the methodology to be employed for the study. In particular, the research design, justification of the mixed methods methodology, the research setting, the data collection instruments and procedures, the participants, and data analysis.
A mixed methods design - incorporating quantitative and qualitative methodological elements - will be used in this study. As defined by Johnson and Onwuegbuzie (2004), mixed methods research is a "third wave research movement" that combines both quantitative and qualitative techniques, concepts, language, and methods in one integrated study (p. 15). Mixed methods research aims at drawing the strengths of both qualitative and quantitative paradigms to provide a comprehensive and thorough investigation of a topic or phenomenon. In this manner, mixed methods research has the effect of harmonizing the rupture of the more dominant research paradigms without replacing them. It capitalizes on the strengths and minimizes the weaknesses of qualitative and quantitative research (Tashakkori & Teddlie, 2003). The mixed methods approach is guided by a pragmatic view of knowledge (Creswell, 2003) focusing on the assertion the truth is "what works" (Howe, 1988). Proponents of mixed methods research select units of analysis and variables that can sufficiently answer the research questions developed (Tashakkori & Teddlie, 1998). The pragmatic view supposes that quantitative and qualitative methods are mutually compatible; no paradigm is better than the other. The idea of mixing both methods is that by combing numerical and textual data, the integration of results is capable of explaining a phenomenon more fully. Hence, the reason behind mixing is that as stand-alone paradigms, both qualitative and quantitative methods could not sufficiently capture the details of a problem of phenomenon, as in the case of a complex educational issue such as truancy. Many research theorists believe that when both methods are combined, quantitative and qualitative methods become complementary and the mix is capable of providing a more thorough and complete analysis (Green, Caracelli, & Graham, 1989).
Considering the qualities attributed to qualitative and quantitative methods more closely, it is easy to view how these paradigms may be irreconcilable. The main line of argumentation of the incompatibility thesis referred to the difficulty of the researcher having a "schizophrenic position" toward the division of labor between quantitative and qualitative methods (Bergman, 2008). The quantitative approach to research makes positivist assumptions in knowledge creation and focuses on measuring the causality of specific variables and testing of hypotheses and theories through statistical tools. The role of the researcher in quantitative studies is detached from the study itself. The researcher is isolated from the subjects of the study and relates casually and objectively. Positivist assumptions require rigor and objectivity in findings; a strict adherence to the scientific method is applied to establish this (Creswell, 2003). Conversely, qualitative researchers assume the inquiry is inherently value-laden (Denzin & Lincoln, 1998). Qualitative inquiry focuses on the process of discovery and how experience is created and the meaning-making that occurs in the process. On the other hand, quantitative studies focus on measuring causal relationships and proceeds in a value-free framework of inquiry (Denzin & Lincoln, 1998). Qualitative inquiry ultimately "produces an understanding of the problem based on multiple contextual factors" (Miller, 2000, p. 14). The primary difference of both paradigms is summed up by Makyut and Morehouse (1994):
The two paradigms are based on two different and competing ways of understanding the world.. which are related in the way research data is collected (words versus numbers) and the perspective of the researcher (perspectival versus objective) and discovery versus proof . (pp 2-3)
This study will use survey data for the quantitative phase; it will then emerge as a qualitative study by using interviews as method for data collection. The interviews will help capture the issue of truancy by reporting findings that will not measure numerically but provide rich and thick descriptions that will add breadth to the statistical results of the quantitative phase. This mixed methods study will look at the effectiveness of the discipline programs in an alternative middle school, the factors that predict truancy among middle school students, and the ways which behavioral and situational variables can be manipulated to improve truancy rates. Quantitative and qualitative methods of data collection will be used such as a web-based questionnaire, interviews, and review of school records.
In deciding the research design for this mixed methods study, three issues were considered: priority, implementation, and integration (Creswell et al., 2003). Priority determines which method is given more emphasis - whether qualitative or quantitative. Implementation determines the process of data collection and analysis - whether it will proceed sequentially (one stage then to the next) or concurrently (parallel stages). Integration means how the results of the two phases of the research process is mixed and connected to one another.
Sequential explanatory design
This study will use the sequential explanatory design (Creswell, 2003; Creswell et al., 2003). The sequential explanatory design is a two-phase design which begins with the collection and analysis of quantitative data which is subsequently followed by the qualitative phase. The first phase will include a web-based survey questionnaire among middle school students which will be subject to descriptive statistics, chi-square, and multiple regressions for analysis. The goal of the quantitative phase is two-fold and in response to the first two research questions, "What are the current behavioral and situational factors that are predictive of truancy in middle school students of alternative schools?" and "How effective is the discipline program within alternative schools for targeting truancy in middle school students with respect to their behavioral and environmental needs?".
The second phase of the study follows through or connects to the results of the first quantitative study. It will involve individual semi-structured interviews of school counselors which will be subject to qualitative data analysis to generate themes and descriptions of data. The goal of this phase is to build explanations on the results generated by the quantitative study; in particular, it will explain why some of the behavioral and situational variables correlated with truancy more significantly than others. Moreover, the interviews will also gather recommendations on how these predictive variables can be manipulated so that truancy rates in middle school students could be reduced. The qualitative data from the interviews will refine quantitative data and explore the phenomenon of truancy intervention more comprehensively.
Follow-up explanations model. The explanatory mixed methods design has two variants: the participant selection model and the follow-up explanations model. They differ in the manner they connect the results of the two phases of the study. The participant selection model emphasizes on how participants can be selected while the follow-up explanations model intends to examine the quantitative results in a more detailed manner. For this study, the follow-up explanations model is chosen because the objective is "to explain and expand on quantitative results" (Creswell et al., 2003, p. 43). In this model, the researcher will identify the findings from the quantitative phase that need to be explained further. For instance, unexpected results and significant differences in the causal relationships of some of the variables with truancy can be included in the interview phase. Explanations and clarifications will be solicited from persons who can best shed light; in this case, the school counselor.
Priority. Because of the nature of this sequential explanatory design and the follow-up explanations model chosen, priority is given to the qualitative phase because it will provide the more in-depth explanation of the effectiveness of the discipline-based intervention in the alternative middle school in reducing truancy rates. The results of the quantitative phase are connected to the qualitative phase. Integration of both phases will come in the write-up of the complete study.
Variables in the Quantitative Analysis
The research questions to be answered in the first quantitative phase are: "What are the current behavioral and situational factors that are predictive of truancy in middle school students of alternative schools?" and "How effective is the discipline program within alternative schools for targeting truancy in middle school students with respect to their behavioral and environmental needs?".
These questions determine the set of variables which will be studied in the first quantitative phase. From the theoretical literature on truancy, several behavioral and situational variables have been identified as having correlative relationships on truancy. Behavioral factors are personal characteristics of the student such as substance abuse, ignorance of attendance laws, mental health disorders, lack of friends, lack of motivation, lack of proficiency in English, boredom, loss of interest, passive views about education (Baker et al., 2001; Munoz, 2001; Clement, Gwynne, & Younkin, 2001). Situational variables that may influence truancy include economic factors, lack of affordable transportation, housing, and single-parent homes (U.S. Department of Education, 1996; Hendricks et al., 2010). The set of variables for this study are detailed below.
I. Independent Variables
Lack of ambition
Lack of attachment to school
Poor peer relationships
Lack of self-esteem
Mental health problems
Substance abuse (alcohol, drug)
Attitudes towards education
Negative peer influence
Family structure (single parent)
Proximity or exposure to violent neighborhoods or homes
These variables will be measured using a web-based survey questionnaire.
II. Dependent variable: truancy
This variable will be measured by computing the number of unexcused absences divided by the amount of days required for attendance. Data will be collected from official school records.
To answer the second research question, effectiveness of discipline programs in alternative middle school in reducing truancy will be based on a 5-year examination of school attendance in the target school.
The target school is part of a network of alternative schools that cater to the needs of students located outside the traditional setting. The DeKalb Transition Academy serves grades 6-8 students who have been released from juvenile detention or have been referred by the DeKalb School System Disciplinary Action Review Committee. Depending on the releases issued, the population is estimated to be 110-140 students in one semester (DeKalb Transition Academy, 2011). A student typically stays in the academy for 90 days (1 semester) before referred back to his or her home school. The DeKalb Transition Academy is classified as an alternative school that uses a community-based model of intervention. The school collaborates with families, schools, the courts, and the community to successfully influence academic outcomes of students with unique needs such as juvenile offenders and chronically truant students.
The target population of this study will be in two groups: 1) middle school students (quantitative), and 2) school counselors and social workers (qualitative). The first target population will be composed of students from grades 6 to 8 in the target school who are classified, per their school attendance, as "truant." Depending on the analysis from the school records, around 50 students in the target school will qualify as participants for this study.
During the second qualitative phase of the study, the target participants are school counselors working in alternative schools within the DeKalb County Schools System. They are school counselors who have extensive experience working with interventions focused on reducing truancy rates among students. Six school counselors will be invited to participate in this phase.
Data Gathering Procedure
Following the work of Johnson and Onwuegbuzie (2004), the data gathering process for this mixed methods study will proceed in eight steps, namely:
Identifying the research question;
Determining the appropriateness of the mixed method design for research;
Deciding the particular mixed method model and design;
Interpretation of data;
Legitimation of data;
Drawing conclusions and final write-up of findings. (p. 21)
Phase I - Quantitative
Participant Selection. The first phase of the study will be quantitative in nature. It will focus on identifying the behavioral and situational factors that predict truancy among middle school students. Employing a cross-sectional survey design, the study aims to collect data at a particular point in time (McMillan, 2000). The participants of this study will be around 50 middle school students enrolled in an alternative school in the DeKalb County Schools System. The participants are those classified as "truant" students based on recorded absence. In order to identify which factors predict truancy among middle school students, participants will be grouped as a) truant, and b) non-truant. In this manner, we can pinpoint the behavioral and situational predictors of truancy that are not present in students who are not classified at truant. Identification of students classified as "truant" will be based on school attendance records and referrals from the school counselor. More specifically, using the definition in Hendricks et al. (2010), students considered "truant" are "students with attendance less than 90%, who were underachieving, marginalized, or experiencing transitional difficulties; had a supportive family or adult sponsor; had no chronic documented health concerns; were without chronic suspensions; and were without significant juvenile offending involvement" (p. 176). Both students and parents will be given an invitation to participate in the study. A consent form will be provided to the parents or guardians of students before the actual conduct of the study. Students who return signed parental consent forms will be considered official participants. Participation will be strictly voluntary and can be stopped at any given time.
Quantitative instrument. The quantitative phase of this study will involve a) survey questionnaire, and b) school records. Data from these sources will be used to answer research problems 1 and 2. Survey data will help identify the predictors of truancy among middle school students. School records particularly those that relate to school attendance will determine the effectiveness of the discipline programs being implemented to reduce truancy.
To answer the first research question, the primary instrument will be a self-report questionnaire which contains items that relate to some of behavioral traits exhibited by the participant as well as the environmental factors that surround the participant which may cause truancy. A self-developed questionnaire will be used to gather quantitative data. It will contain multiple format of responses such as multiple choice questions, dichotomous "yes" or "no" questions, self-assessment items, and items measured via a 5-point Likert-type format. The questionnaire will have a total of 20 questions to properly cover all the quantitative variables to be studied. In order to guarantee the content validity of the self-developed questionnaire, a panel of experts from the University will serve as evaluators.
The questionnaire will have four sections. The first section will gather responses related to their experiences with discipline programs in the alternative middle school. The second section will consist of self-assessment items that will determine the types of behaviors they feel they possess and their attitudes towards school. The third section will consist of items pertaining situational variables they face as middle school students that may influence their school attendance. The fourth section will be composed of demographic questions such as the student's 1) age, 2) gender, and 3) ethnicity. Some of the questions in the instrument will be open-ended such as the inclusion of an item "Other" so that students can specify the correct answer. The option of "N/A" or "not applicable" will also be provided. The surveys will be administered via the Internet. The service SurveyMonkey.com will be used. Upon giving consent, students will be provided a URL to the survey. One of the biggest advantages of a web-based survey is the ease in which the data can be transferred to Excel or SPSS for tabulation and for data analysis.
To answer the second research question, school data will be collected and analyzed to determine the effectiveness of discipline-based programs to reduce truancy rates in the target school. Records will be accessed after obtaining permission from the school. Administrative data will be collected from the target school, specifically school records spanning five years from 2005-06, 2006-07, 2007-08, 2008-09, 2009-2010 school years. The specific data to be gathered are 1) the number of days attended, 2) the number of excused absences, and 3) the number of unexcused absences during each school year. The standard truancy rates will be computed by taking the number of unexcused absences that occurred in any given period (e.g., 30-day period) and dividing it by the number of school days during that period. This will reveal whether the discipline programs being implemented in the alternative school to reduce truancy has led to actual improvement in school attendance and a decrease in unexcused absences.
Reliability. In order to establish reliability, the self-developed instrument will be pilot tested to around 50 students or 5% of the total student population. Participants will be randomly chosen. The pilot test is a way of validating the instrument by measuring its internal consistency reliability. In order to test reliability, Cronbach's alpha coefficient will be computed (Gliem & Gliem, 2003). The standard or normal range for internal consistency is a Cronbach's alpha between 0 and 1. The internal consistency of an instrument is higher when the coefficient computed is closer to 1. The interpretation of the coefficients are "_ > .9 - Excellent, _ > .8 - Good, _ > .7 - Acceptable, _> .6 - Questionable, _ > .5 - Poor, and_ < .5 - Unacceptable" (George & Mallery, 2003, p. 231). The instrument will be revised accordingly depending on the results of the pilot test.
Quantitative Data Collection. Formal letters to obtain permission to access school records from the target school will be made. After permission is granted, the researcher will solicit the help of the school administrator and the school registrar for the data required. From the data on school attendance, the "truant" and "non-truant" groups will be classified based on the number of school days missed and number of excused and unexcused absences. One week before the online survey URL is operational, the researcher will send an email notification to participants asking for their cooperation and informing them of the importance of the study. A low response rate is expected of web-based surveys. In order to secure a higher rate of response, a two-phase follow up will be made. Target participants who have not completed the survey will be sent a reminder email 1) five days after the URL is provided, and 2) two weeks after.
Data Analysis. Descriptive statistics will be used to summarize the responses in the survey and to present them in tabular form.
RQ1. In order to analyze survey data for the research question "What are the current behavioral and situational factors that correlate with truancy in middle school students of alternative schools?", chi-square and multiple regression analysis will be conducted. The chi-square values determine whether the independent variables (behavioral and situational factors) have a significant relationship with the dependent variable (truancy rates). Multiple regression analysis is the appropriate statistical method to measure relationships across two or more independent variables.
RQ2. A t-test for dependent means will measure whether there is a significant improvement in the standard truancy rates of the target school for five school years.
All statistical analysis of the quantitative results will be conducted with the help of Statistical Package for Social Sciences software (SPSS), version 11.0.
Phase II - Qualitative
Participant Selection. For the second qualitative phase of this study, the third research question will be answered. In order to determine how behavioral and situational factors can be manipulated in order to reduce truancy rates among middle school students, in-depth interviews of school authorities will be made. To this end, purposive sampling will be used in order to solicit more information on the phenomenon being studied (Creswell, 2006). Participants for the second phase will six school counselors from the DeKalb County Schools System. According to the sequential explanatory participant selection model, participants for the second phase were selected for a follow-up, in-depth, qualitative phase after analyzing the quantitative data in the first phase of the study (Creswell & Plano Clark, 2007). In the participant selection model the emphasis is on the second, qualitative phase of the study to gain a better understanding of the predictive variables of truancy among middle school students and how to address it.
Qualitative Instrument. The instrument used in the second phase of this study will be a semi-structured interview protocol. It will consist of around ten to fifteen open-ended questions that will focus on the role of discipline programs in addressing behavioral and situational predictors of truancy among middle school students. The content of the interview protocol, being an extension of the statistical results of the quantitative phase of the study, will be based on the results of the previous quantitative phase. The questions will emphasize on truancy and the significant variables that predict absence and non-attendance among middle school students. The protocol will be pilot tested on three school counselors.
Data Collection. The second, qualitative phase in the study will focus on explaining the results in the first, quantitative phase. The primary technique will be conducting in-depth semi-structured personal interviews with three (3) school counselors and three (3) social workers from the DeKalb County Schools System. Triangulation of different data sources is important in qualitative study (Creswell, 1998) so the participants will be requested to bring additional materials or sources that could verify some of their assumptions or statements. In order to ensure that they are prepared for the interview, the researcher will send the interview questions ahead of the scheduled time. A consent form will also be attached that will explain the purposes and objectives of the study and that the entire interview will be tape-recorded and transcribed verbatim. Interviews will be done over time rather than in one single instance. This will provide the participants time to review, correct, or clarify the contents of the transcribed interview.
Qualitative Data Analysis. In the qualitative analysis, data collection and analysis proceed simultaneously (Merriam, 1998). In the second, qualitative phase of the study, the text data obtained through the interviews and other relevant documents will be coded and analyzed for themes with the help of the Qualitative Software and Research (QSR) N6, software for qualitative data analysis.
The steps in qualitative analysis (Creswell, 2006) will include:
preliminary exploration of the data by reading through the transcripts and writing memos;
coding the data by segmenting and labeling the text;
using codes to develop themes by aggregating similar codes together;
connecting and interrelating themes; and
constructing a narrative. (p. 132)
The first step in qualitative analysis is transcribing the interviews. Recorded interviews will be played back and typed out by the researcher. The second step is to read the data line-by line and make sure the transcriptions are accurate. The third step is the coding process, writing codes in the paper margin, then organizing the codes into categories in an excel spreadsheet using terms from the actual language of the participants. The interview responses were coded to see if any differences emerged. The fourth step will use codes to develop larger themes, which will then become the major headings in the qualitative results section of the study. In step five, the themes will be represented through quotes in order to substantiate the themes and to represent what the study found. The sixth step will involve the interpretation of the data. During this step the meaning from the qualitative data will be connected to the literature on factors that influence truancy among middle school students.
Establishing Credibility. In judging the qualitative study, the criteria are different. When in quantitative design, validity and reliability are established, the qualitative researcher is more concerned with credibility (Golafshani, 2003). Credibility refers to the believability of the findings through the coherence and utility of the gathered data (Eisner, 1991) and trustworthiness (Lincoln & Guba, 1985). This is achieved by subjecting the data through a verification process as opposed to statistical determination of reliability and validity measures. The findings of a qualitative study are limited to a particular context and could not be generalized elsewhere. The interpretation of qualitative data is also at risk for subjectivity of the researcher's personal biases and positions on an issue. The manner of establishing rigor in qualitative study means that the risk for bias and subjectivity is minimized (Creswell, 2003).
Validating the findings of qualitative study is ensuring that the information presented matches reality (Merriam, 2009). The following forms of ensuring credibility will be used in this study: 1) triangulation, 2) member checking, 3) the use of rich, think descriptions, and 4) an external audit. Triangulation will be done by using multiple sources of data such as interviews, documents, and secondary sources. Member checking is ensuring the accuracy of the representation of data by allowing interviewees to check and if necessary, correct the interview transcripts. Rich and thick descriptions will allow for a deeper understanding of the phenomenon. External audit will be performed by a qualitative research expert who will review the method and the findings of the study and give feedback thereafter (Creswell, 2003; Creswell & Miller, 2002).
Strengths and Weaknesses of the Sequential Explanatory Mixed Methods Design
The advantages and limitations of mixed methods design are well discussed in the literature (Creswell, 2002; Creswell, Moghaddam, Walker, & Harre, 2003). The strengths of this particular research mixed method paradigm are:
It is relatively easy to implement for a single investigator since the process is sequential and will be conducted from one stage to the next. It is not as taxing or consuming as the concurrent model.
Sequential explanatory mixed methods design is helpful in understanding or shedding light to the results of quantitative study more comprehensively (Green & Caracelli, 1997)
This particular design is particularly helpful when the results from the quantitative phase do not match theoretical expectations (Morse, 1991).
On the other hand, the weaknesses of the sequential explanatory mixed methods design are:
Compared to pure qualitative or quantitative designs, this mixed method model is lengthy and more time-consuming.
The first quantitative phase may not generate the results expected or not show significant difference.
Research Permission and Ethical Considerations
Ethical issues will be resolved for each phase of this mixed methods study. To comply with Institutional Review Board (IRB) regulations, permission to conduct the two phases of the study will be secured. The researcher will complete The Request for Review Form and personal and professional details as well as project information, including title, type, funding sources, and subjects.
Since the subjects of the first quantitative phase of the study are middle school students aged below 18, parental consent forms need to be obtained. To this end, the researcher will develop a consent form that states how the study will protect the confidentiality and rights of the subjects. A similar form will be attached to the web-based survey to prove that the requirements for ethics have been complied. The participants will be guaranteed anonymity. Each returned questionnaire will be numerically coded so that their names are not divulged and the information is confidential. The letter will clarify that the participation of students is voluntary and that confidentiality will be guaranteed in data collection and presentation. For the personal interviews, the transcripts will not reveal their names; only fictitious names will be used in describing and reporting the results. The volume of data produced such as interview tapes and transcripts will be stored in a secure metal file cabinet and will be destroyed after publication of the study. It will also be explained to the participants that while the summary data may be accessible by the professional community, it will not be traced back to them as respondents.
Research bias is acknowledged and considered in designing this study. Bias may exist at different levels and degrees but verification processes have been identified to minimize the risk of bias. The interpretation of the results from the qualitative phase may be prone to subjectivity because of the position and passion of the research in advocating for educational reforms particularly in truancy intervention. While not directly employed in an alternative school, the researcher has served in an administrative capacity at the target school and is deeply committed to the needs and potential of at-risk students. The credentials and possible conflicts of interest that the researcher may have will be fully disclosed.
The Role of the Researcher
The involvement of the researcher will be different for the two phases of this mixed methods research. During the first phase, the researcher will be involved in data collection, grouping the participants into truant and non-truant groups, administering the survey questionnaire by putting up the URL that links the participants to the web-based instrument. For the second, qualitative phase of this study, a more participatory involvement is expected of the researcher because of "sustained and extensive experience with participants" (Creswell, 2003, p. 184) as well the researcher's experience and personal involvement with the topic of truancy. In this phase, the researcher will conduct the interviews with school counselors.
The researcher is currently serving as the 8th grade assistant principal at Lithonia Middle School. Her experiences in truancy intervention have been extensive and will be disclosed fully. Having spent 18 years in the DeKalb County Schools System, the researcher has been involved in truancy improvement programs in alternative schools. This includes informing the parents about the truancy improvement program, consulting parents and students experiencing problems with truancy, and coordinating efforts with teachers in regards to problems related to the truancy improvement program. The researcher has also handled discipline referrals in her previous roles in the alternative school.
While no longer directly connected to the target school, the researcher still belongs to the same schools system as the participants in the second phase of the study. The possibility that the researcher is acquainted with some of the participants is possible. Moreover, the researcher's experience and involvement with students and teachers form the beliefs and assumptions she still holds today. This may influence the researcher's interpretation and control over the study. Therefore, care and effort will be directed at minimizing the risk and ensuring the objectivity of the study. It is pointed out that while the researcher belongs to the same county, the locale is not within the researcher's own school or classroom. The target school was selected purposefully.
Nonetheless, the study's design will ensure that bias and subjectivity will be eliminated as much as possible. In this regard, four forms of verification are proposed in order to enhance credibility. This includes triangulation, member checking, rich and thick case descriptions, and an external audit.