Chapter 4 Example: Impacts of Bullying on Students’ Performance Rate and Mental Health

Overview of this Chapter 4 Dissertation Example

This Chapter 4 of the dissertation example is developed by Best Dissertation Writers. Chapter 4 examines the relationship between bullying and school absenteeism, a topic of increasing research interest. The study aims to contribute new knowledge to this field, focusing on students aged 18 and above in rural settings. It explores the correlation between bullying and increased absenteeism rates, as well as the impact of peer social support on absenteeism. The chapter addresses conflicting views in existing literature, with some studies supporting a correlation between bullying and absenteeism while others oppose it. The research is structured around specific hypotheses and questions, seeking to either support or challenge previous evidence on these relationships. This investigation is crucial given the rising incidence of bullying in schools and its potential effects on student performance and mental health.

Chapter 4: Findings

Bullying as an Important Problem in School

Bullying is among the key factors that affect that performance rate and mental health of the students in the contemporary learning environments (Grinshteyn and Yang 2017). The relationship between bullying and school absenteeism has been widely researched in the recent years, with different scholars providing varying views, as some of the support the correlation (Gruber and Fineran 2017) while other oppose it (Pampati et al. 2020). Therefore, the present study was focused on generating new knowledge to support or disapprove the previous evidence about the relationship between bullying and absenteeism. As a result of the increasing cases of bullying in school, there has been need to develop strategies for addressing the problem. For that matter, this study also focused on assessing impacts on peer social support on absenteeism rates in school. The primary focus of this study was to assess the existence of relationships between bullying and increased rate of absenteeism among the school students aged 18 years and above living within a rural setting. Additionally, the correlation between the peer social support and the rate of absenteeism was addressed. The chapter is organized into different subsections, mainly based on the research hypotheses and questions that were to be answered using the generated outcomes.

Validity and Reliability

Meghanathan (2015) noted that both reliability and validity deal with the trust or faith that an individual has in the findings as well as any type of conclusion that is made from the generated outcomes. Maric et al., (2015) defined reliability as the similarity of the study outcomes provided by the independent but comparable measures of the same object. On the other hand, validity is the level of accuracy possessed or expressed by the methods used for measuring the variables and their ability to measure what is expected of them (Meghanathan, 2015). In the present study, different statistical approaches where used to ensure that the issues associated with reliability and validity were effectively addressed. After successfully formulating the hypotheses to be tested, operationally defining all the variables whose relationships are to be assessed, as well as specifying the levels of measurements and the statistical analyses to be used, the researcher no has the responsibility of addressing other issues associated with validity and reliability.

Binding is another approach was used in this study to promote internal validity. Specifically, the behaviors of students included the study were observed in their natural settings in order to prevent them from developing prior perception that their bullying activities were being investigated. As advised by Maric et al. (2015), following a specific protocol during the execution of study activities may also help in improving the internal validity of the study. In this study, the researcher followed a specific protocol for participant recruitment, data collection and analysis. All of the participants were treated using the measures as an approach for limiting the occurrence of bias as well as improving the overall outcomes of the study.

Furthermore, the external validity of the study was also improved. Specifically, the generalizability of the study outcomes was improved by collecting data from large sample size hence leading to the promotion of population validity. In line with the explanations by Meghanathan (2015), population validity often depends on the population choice and the extent to which the study sample mirrors the actual population. As a result of the high cases of bullying among the high school students, the present study included only data from high school students; hence making the generated data to be generalized to the general high student population aged 18 years and above. However, the outcomes could not be generalized on the general high student population because the present study only included students aged 18 years and above, which are often the minority in most of the high school settings. 

Hypothesis testing was one of the primary tests that were conducted in order to determine the relationship between the bullying and school absenteeism. The approach was also used for ensuring the validity and reliability of the collected data and study outcomes. The assumptions used during the hypothesis testing process in this case include (1) the hypothesis must be declarative, (2) must be able to express the correlation between two or more variables and (3) must be capable of being empirically tested (Meghanathan, 2015). All of these assumptions generally imply that the variables included in the hypotheses must be measurable as an approach for testing the effectiveness and appropriateness of the collected data as well as the generalizability of the study outcomes (Maric et al., 2015). As an approach for meeting the requirements for the first assumption, the present study originally declared (guided with the evidence from literature) that bullying is a key influencer of high rates of school absenteeism and that peer social support is an important intervention for addressing truancy or absenteeism in school. The second assumption was met in the present study as it was focused of assessing the relationship between bullying and school attendance, and the relationship between peer social support and reduced bullying and absenteeism incidences.

The main threat to internal validity that was experienced was the selection bias, especially during the requirement of the participants. Inclusion and exclusion criterion were used in order to prevent the realization of adverse effects of this threat. Furthermore, the reliability aspect of the research was heightened using adequate number of questions for assessing competence as well as ensuring that all of the participants were familiar with the assessment user interface. The use of OBQ survey data and the CASSS survey helped in promoting the psychological validity and soundness as both the tools had the ability of achieving their intended goals. The questionnaires were able to measure the impacts of bullying and social support on the absenteeism rates among the learners.

Generally, external validity represents lack of interaction between the study variables, the subjects and the environment (Meghanathan, 2015). Therefore, the researcher must define the specific characteristics which make the participants unique that prevent the generalization to the other subjects or settings. In the present study, external validity was achieved by applying inclusion and exclusion criteria during the selection of the participants, data collection and analysis. Furthermore, the participants used in the final study had similar characteristics to those included in the pilot study so as to prevent the occurrence of external validity issues. The decision to use this approach was motivated by the arguments by Razali and Wah (2015) that using volunteers in the pilot study may negatively impact the external validity of the final study because some of the volunteers might have characteristics which are different from those participants to be included in the final study. An important reason for ensuring external validity is to promote the generalization of the study outcomes. Based on the fact that the present study included data from a large study area and large number of participants, its outcomes could be easily generalized to the large student population, beyond the Virginia School District.

Results

Hypothesis 1: There Is Significant Relationship Between Being Bullied and School Absences

Table 1 shows the Pearson correlations between the study variables. As shown, the correlation between Bullying and Absences was not significant (r = 0.192; p = 0.201). Therefore, the null hypothesis which states that “there is no significant relationship between being bullied and school absences” was accepted. Prior to the assessment of the collected data, the study assumed that there is a positive correlation between being bullied and school absences. Nonetheless, contrary findings were realized as the null hypothesis was accepted.

Table 1: Pearson Correlations Between Study Variables

	Peer Support		Bullying*
	R	p	r	p
Absences*	0.145	0.338	0.192	0.201
Peer Support			-0.056	0.710

* Logarithms were taken to normalize the distributions

Hypothesis 2: There Is Significant Relationship Between Peer Social Support and School Absences

As shown in Table 1, the correlation between Peer Support and Absences was not significant (r = 0.145; p = 0.338). Therefore, the null hypothesis which states that “there is no significant relationship between peer social supportand school absences” was accepted. Guided with the information from the previous literature, the study assumed that there might be a positive correlation between peer social support and school absences; in that the availability of peer social support would reduce the rate of school absences. There are five important assumptions that influence the application of Pearson’s r and they include two variables being continuous, two variables being paired, there is a linear relationship between the variables, with no significant outliers. The distributions of both variables are assumed to follow a normal bell-shaped curve. Figure 1 displays histograms of the Absences and Bullying distributions. Both variables appear to have significant skewness to the lower ends of the distributions.

Figure 1: Histogram of Absences and Bullying Distributions

Figure 2: Histogram of Peer Support, Which Appears to Be Normally Distributed

Demographic Characteristics of the Participants

Table 2: Demographic Characteristics of The Participants

Characteristic	Category	Frequency	Percentage
Gender
	Male	3	6.5%
	Female	15	32.6%
	not specified	28	60.9%
Ethnicity
	African-American	1	2.2%
	Asian	1	2.2%
	Caucasian/White	34	73.9%
	Hispanic	4	8.7%
	Other, ethnic group	6	13.0%

Demographic information of the surveyed participants included ethnicity as follows: 1 African American (2.2%), 1 Asian (2.2%), 34 Whites (73.9%), 4 Hispanics (8.7%) and 6 (13.0%) from the remaining ethnic groups within the study population. Despite the fact that the data were collected from 46 participants, only 18 of them specified their gender. Out of that number, 15 were female and three were male. Due to the small number of respondents who provided their gender, analyses could not be generated to compare the variables under investigation (bullying, social support and absenteeism) by the gender orientation of the participants.

In order to substantiate the level of skewness in the distributions, the normality of the variables was assessed using z-scores formed by dividing skewness by the standard error of skewness. A z-score within +/- 3.29 is indicative of a normal distribution (West, Finch, & Curran, 1995).  The results are presented in Table 2. The row labeled “Bullying” contains data that were collected by using the OBQ survey, while the “Peer Support” row is a presentation of data that were collected through the use of CASSS survey. Furthermore, the “Absences” row is a presentation of the absenteeism data, which were retrieved from the school division’s archived data storage system. Based on the z-scores shown in the last column, both Absences and Bullying exhibited substantial skewness (Absences z = 6.21 and Bullying z = 5.03). Normalizing transformations were applied to both variables according to recommendations provided by Tabachnick and Fidell (2013). Both variables were successfully normalized using logarithms, and the logarithms were used in all subsequent analyses. Peer Support was confirmed to be normally distributed (z = -0.83).

Table 3: Descriptive Statistics of The Collected Data (n = 46)

Variables	Mean	SD	Minimum	Maximum	Skewness	SE	z
Absences	4.76	6.13	0	26	2.18	0.35	6.21*
Bullying	69.52	14.31	52	119	1.76	0.35	5.03*
Peer Support	49.17	12.42	12	72	-0.29	0.35	-0.83
Normalized variables (logarithms)
Absences	0.93	1.29	-1.39	3.27	-0.34	0.35	-0.98
Bullying	4.22	0.18	3.95	4.78	1.15	0.35	3.29

* skewed distributions

Evaluation of the Findings

Although no significant relationships were observed between the school absences and either bullying or peer social support, a multiple linear regression analysis was conducted to ascertain the combined predictive value of both bullying and peer support on the number of school absences. Based on the explanations provided by Schroeder et al., (2016), an R squared value above 0.4 is an indication of a strong relationship, a value between 0.2 and 0.4 is an indication of moderate relationship, while a value below 0.2 is an indication of weak correlation. As shown in Table 4 below, the R squared value was very low (R² = 0.06). Based on the analysis of variance of the model, the prediction of Absences using Bullying and Peer Support was not significant (F (2,43) = 1.40; p = 0.258). Furthermore, the beta weights for both predictors were not significant (β = 0.20; t = 1.36; p = 0.182 for Bullying and β = 0.16; t = 1.05; p = 0.298 for Peer Support).

Table 4: Regression on Absences Using Bullying and Peer Support as Predictors

	R	R2	F	df	p	β	t	p
Model	0.25	0.06	1.40	2,43	0.258
Predictors
Bullying						0.20	1.36	0.182
Peer Support						0.16	1.05	0.298

Note. Logarithms were taken of Absences and Bullying to normalize the distributions

Hypothesis 1: There Is Significant Relationship Between Being Bullied and School Absences

Correlational analysis of the collected under bullying and school absences led to the generation of r = 0.192; p = 0.20. The generated outcomes showed that there is no significant relationship between being bullied and school absences; hence leading to the acceptance of the null hypothesis since the generated p value was higher than 0.005 which is an indication that there is no strong correlation between the two variables. Even though the produced r value was positive, it could not be used to fully support the probable correlation between being bullied and school absences. The positive r-value of 0.192 shows that there is a positive correlation between bullying and absenteeism. However, the magnitude of such correlation is limited because the generated p-value was higher than 0.005 (0.20), which is an indication that there is a weak relationship between the two variables.

The outcomes therefore contradict the arguments by Juvonen and Schacter (2017) and Zych, et al. (2017) that bullying is one of the primary factors that influence increasing rates of school absenteeism among the high school learners. In those studies, a general conclusion was made that bullied students do often absent themselves from the school because of the fear of being bullied again or lack of interest in the schooling process. However, Pampati et al. (2020) and Baldry et. al. (2017) on the other hand identified bullying as an important factor that increases the chances of a student to absent his or herself from school and that being bullied is not a guarantee that the students would be absent from school. Therefore, the outcomes from this study specifically agree with the findings from the research works by Pampati, et al., (2020) and Baldry, et. al., (2017) as they both identify bullying as a factor for increasing school absenteeism and that not all the bullied students will be absent from school.

Hypothesis 2: There Is Significant Relationship Between Peer Social Support and School Absences

The present study rejected the alternative hypothesis which states that “there is significant relationship between peer social support and school absences” because the produced p value was higher than 0.005. Precisely, a p value of 0.338 and r value of 0.145 were produced from the analysis. Even though the correlational analysis of the data collected under peer social support and school absences produced positive correlational value, the outcomes could not be used to make a final decision that the exitance of peer social support may help in addressing issues related to school absences because of the higher score of p value that was obtained. Therefore, the outcomes can be used to argue that peer social support is one of the strategies for reducing school absenteeism and that it is not a guarantee that those students who are offered such services would not absent themselves from school.

The findings from this study agree with those from the previous studies by Bennett et al. (2018), Green et al. (2019), and Grinshteyn and Yang (2017), which identified peer social support as an important tool for educating students on the impacts of bullying and managing the increasing cases of bullying. Even though Grinshteyn and Yang (2017) and Green et al. (2019) also identified peer social support as in effective strategy for addressing school absenteeism cases, they failed to identify the precise circumstances under which the strategy can be effective. With reference to the evidence presented in the studies by Grinshteyn and Yang (2017) and Smalley et al. (2017), a strong positive relationship between peer social support and school absenteeism cannot be maintained because the level of outcomes from this strategy is influenced by different factors such as ability of the students to commit themselves in applying the advices provided by the support team. Furthermore, the level of professionalism practiced by the individuals tasked with the role of executing the peer support service may also impact the quality of outcomes to be registered. Therefore, the outcomes from the hypothesis 2 analysis agree with some previous literatures and also contradict some of the published findings concerning the nature of relationship between peer social support and school absences.

Summary

Chapter 4 of this dissertation involved comprehensive presentation of results obtained from the data analyses process. The statistical analyses conducted included descriptive statistics, normality testing, Pearson correlations, and multiple linear regression. The main goal of the chapter was to present results with regard to the two hypotheses proposed in the study. In this regard, both the null hypotheses were accepted. Furthermore, multiple linear regression analyses revealed that there was no significant combined impact of being bullied and peer support on the number of school absences. Chapter 5 will provide critical interpretation of the results presented in the Chapter 4, as well as comparing the results with those from the previously completed studies in similar research areas.