Sustainable Way of Using Nuclear Power and Waste Disposal

¶ … Management Strategy to Utilize Meta-Analysis Technique for Nuclear Energy and Waste Disposal and Create Social Sustainability

A Dissertation Presented using the Meta-Analysis Technique

Komi E Fiagbe Comment by Owner: This is exactly what I wanted to see Komi. Your literature review should also identify the 20 studies that you want to analyze to give your readers an understanding of what you are doing. I want you to email me all three chapters as we move into the next course.Dr. A Christina Anastasia PH-D Chair

[Committee Name], [Degree], Committee Member

[Committee Name], [Degree], Committee Member

This research proposal explores the link between public perceptions of nuclear power, how those perceptions are formed, and what influence those opinions have on energy policy. These issues are important in light of two realities. First, nuclear energy is declining in its share of global energy. Second, nuclear energy offers what might well be the best solution to climate change. Given the threat posed by climate change, it makes sense that nuclear power would be increasing in share, not decreasing. This Research proposal seeks to look at some of the issues facing nuclear power, and how it can overcome these issues to increase share going forward.

[Add a Dedication, if desired [Add Acknowledgements]

Table of Contents

List of Figures

[Add List of Figures here vii

Chapter Three

Research Traditions

There are a number of different research traditions that could be used to understand a subject as broad as nuclear power. Generally, the wide nature of this issue implies that the research methodology utilized, measurements undertaken, and the kinds of people examined differ depending on the study while the topic is the same (Glass, 1977, p.351). The appropriate method, ultimately, depends on a number of different variables.

The availability of hard data will determine whether quantitative methods can be used. What data is available is dependent on the research question itself. In the case of this project, the research question focuses on the future of nuclear power. Nuclear power decisions are made most often at the political level, as a matter of national energy policy, rather than the scientific level. In terms of the science, there is little doubt that nuclear power can be used, albeit with some fairly significant risks (Alic, 2012).

Nuclear power has remained an increasingly controversial issue for a long period of time. The controversy has largely been centered on whether using this type of power as an energy source would be beneficial or harmful to the society and environment. Even though nuclear power is associated with some harmful effects that could be devastating to the environment, there are significant benefits attributed to using it as an energy source.

The controversial nature of this topic implies that a suitable methodology has to be selected for this study in order to generate accurate findings and conclusions. Actually, this research issue entails examining the benefits and harms associated with nuclear power in order to develop management strategies for using nuclear power and waste disposal in a sustainable manner (Alic, 2012). The researcher's ability to effectively examine these issues and develop accurate conclusions for future studies and implementation is directly affected by the research methodology.

Moreover, the broad nature of the research issue implies that the researcher must careful consider the various research methodologies and approaches before selecting a suitable one. In light of these factors and considerations, the most appropriate methodology for this research would be the Meta- Analysis Technique, which is a statistical process that integrates the findings of different independent studies that are regarded as combinable (Egger, Smith & Phillips, 1997, p.1533).

Meta-Analysis technique is an appropriate research methodology for this study mainly because effectively performed meta-analyses promote a more objective evaluation of evidence or selected studies that conventional narrative or literature reviews (Egger, Smith & Phillips, 1997, p.1533). The other benefits of using meta-analysis technique in this research include its provision of more accurate prediction of results and comprehensive explanations of the heterogeneity between findings of individual studies. In this case, meta-analysis will enable the researcher to provide comprehensive explanations of the heterogeneity between the results of several studies regarding the use of nuclear power as an energy source and waste disposal in a sustainable and beneficial manner. However, these benefits would be realized through carefully planning the research project via a comprehensively written protocol that is prepared before commencement of the study.

The suitability of meta-analysis technique for this research is attributed to the fact that it's a collection of methodical techniques for resolving obvious contradictions in research findings. This technique also involves translating results from varying studies to a universal metric and statistically examining relation between research characteristics and findings (Bangert-Drowns & Rudner, 1991).

According to Egger, Smith & Phillips (1997), meta-analysis is a statistical process that entails integrating the findings of various independent studies that are considerably combinable (p.1533). Similar to any other research process, the use of meta-analysis in this study will be characterized by several steps i.e. formulating the issue or problem to be studied, data collection and analysis, and reporting the findings. Meta-analysis technique is also appropriate for this study since it promotes clear definition and analysis of the issue or problem being examined. Actually, natural, biological, and physical sciences usually enable research problems or issues to be clearly defined and evaluated through uniform and commonly accepted methodologies and measures (Wolf, 1986, p.9).

As previously mentioned, nuclear power is a broad topic that has attracted significant attention in the recent past, which has contributed to numerous research and studies on this issue. In essence, the broad nature of the study not only contributes to use of different research traditions but also generate various kinds of studies. The findings of the numerous existing studies on the issue of nuclear power are relatively fragile as research on the topic is increasing at an astounding rate, particularly because of increased climate change in the recent past. Research findings of existing studies on the issue of nuclear power are fragile since they are characterized by irregularities across several factors such as contexts and classes of subjects.

This is usually a major challenge in education research because of the rapid increase in research on dozens of topics (Glass, 1976, p.3). Therefore, the need for meta-analysis technique in conducting this study is relatively clear. Through the use of this technique, the researcher will primarily focus on examining different studies in terms of their information and contradictions before statistically examining their characteristics and findings. This process will be geared towards integrating the results of these independent studies to generate significant conclusions regarding the topic. The use of meta-analysis technique in the study will help generate important information about the issue.

Thus, this is not a paper rooted in the technical sciences, but rather in the social sciences. The matter at hand in the four research hypotheses that are being examined is essentially qualitative in nature. The studies that have been produced are somewhat thin on the ground, there have been a handful of studies, but none so closely related that they can be subject to traditional quantitative analysis. But this does not mean that these studies cannot be analyzed for common traits, nor does it mean that we cannot extrapolate from a grouping of studies truths about the world at large. To be certain, one country's experience cannot fully and smoothly equate to another country's experience, there is always local context, but again, that is inherent in the social sciences. Social sciences work is not conducted in a laboratory, but in the real world. The problem with quantitative studies is that they leave little room for things like context and nuance; only qualitative analysis by an expert practitioner can result in a meaningful interpretation of diverse sets of qualitative and quantitative data, for instance the Meta-Analysis Technique (Ritchie & Lewis, 2003).

However, since qualitative research does not entail creating research hypothesis, the researcher will also utilize quantitative methods in this study. The significance of quantitative research method is that it enables development of several hypotheses that are used in the research process in order to examine the specific issue under evaluation. On the contrary, qualitative research will be utilized since it is more effective in social research.

Given that this study is not based on technical sciences but on social science, qualitative research will be vital. Actually, qualitative research has a longstanding history of providing significant insights regarding social structures, cultures, and behaviors. Moreover, qualitative methods help in development and evaluation of social policy, which makes them effective in social research (Ritchie & Lewis, 2003, p.25).

Type of Meta-Analytic Approach to be utilized

There are a number of critical competencies that a researcher needs to have in order to apply expert analysis to a given subject matter. The research needs to be able to effectively manage information; the organization of the literature review along the individual research questions was instrumental in developing the needed organization to answer these questions. There also needs to be systemic competencies such as research capabilities, the ability to apply knowledge, the ability to adapt to one's set of circumstances and creativity (Verd, 2010).

The field of meta-analysis has different approaches that can be used depending on the suitability of the approach in examining the research issue. These approaches differ on the basis of processes, computations of research findings, and interpretation of these findings. As a result of the existence of different meta-analytic approaches, researchers need to clearly state the approach utilized in the specific study. Glass (1977) argues that academic research and analysis is wide and broadly scattered as evident in the production of hundreds of studies within a short period of time (p.351).

Therefore, research techniques, undertaken measurements, and types of people studies tend to vary from one research to the next even if the research issue or problem is the same. The nature of academic research and evaluation in the past few years as well as differences in these components imply that selecting a suitable research methodology is crucial towards generating accurate research findings and conclusions. Therefore, the researcher will clearly state the type of meta-analytic approach to be utilized in this study based on its suitability in the research. However, the researcher will provide a brief description of the other meta-analytic approach that is commonly used in research projects before elaborating the techniques of the selected approach in light of their suitability in this study.

The two most common types of meta-analytic approaches that are commonly used in researches are Study Effect Meta-Analysis and Classic or Glassian Meta-Analysis. Study Effect Meta-Analysis differs from the other approach on the premise that it has more selective inclusion rules and the study act as the unit of analysis (Shachar, 2008). In this approach, researches with serious methodological mistakes are excluded from the study while one effect size is computed for every research or study.

In contrast, Glassian Meta-Analysis is regarded as the precedent or pattern for conventional meta-analysis and adopts liberal inclusion criteria. In this approach, the study finding is the unit of analysis and allows several comparisons between groups and subgroups on varying criteria. Unlike in Study Effect Meta-Analysis, the effect sizes in Glassian Meta-Analysis are calculated or computed for every comparison.

For this study, the researcher will utilize Glassian or Classic Meta-Analysis, which incorporates various steps including defining questions to be analyzed, collecting data, and examining relations between code study characteristics and results. As previously mentioned this meta-analytic approach applies a more liberal inclusion criterion and considers the study finding as the unit of analysis (Shachar, 2008).

These two components of Glassian Meta-Analysis are important in this study because of the nature of the issue or problem being examined. Since nuclear power is a broad and increasingly controversial issue, liberal inclusion criteria will enable the researcher to examine many studies relating to the topic area through a robust search of the topic area. The liberal inclusion promotes a robust search of issues relating to the topic and enables the researcher to carefully and effectively select studies that provides significant insights to this issue (both published and unpublished)

The use of liberal inclusion criteria for this study is also based on the fact that the controversial nature of nuclear power has contributed to numerous studies being conducted on this issue. The availability of huge literature and data relating to nuclear power implies that a restricted inclusion criteria would limit the ability of the researcher to gain more insights on this issue and hinder the likelihood of generate accurate research findings and/or conclusions.

Secondly, the use of study findings as the unit of analysis is crucial in this research since meta-analysis process will entail making comparisons of research findings and combining research results. If the study is used as the unit of analysis in the way it is carried out in Study Effect Meta-Analysis, the researcher will seemingly be unable to make effective comparisons of study findings. The nature of the research requires research findings to be the premise or unit of analysis so that the researcher can combine research results while making comparisons between study findings.

According to Glass (1976), meta-analysis can be described as the analysis or evaluation of analyses in which statistical analysis of a large collection of the findings of individual studies is carried out in order to integrate these findings (p.3). Based on this definition, the unit of analysis is meta-analysis is the findings or results of individual studies utilized in a research project. Since this study focuses on analysis of findings of individual studies in order to integrate them, Glassian Meta-Analysis will be utilized as the meta-analytic approach.

Glassian Meta-Analysis involves the use of statistical methods to conduct integrative analysis since meta-analysis differs from conventional narrative research methodologies on the basis that it is increasingly statistical and quantitative (Glass et. al, 1980, p.1).

As a result, meta-analysis approaches are simply statistical techniques that are appropriate and applicable to the process of integrating study findings. Glassian Meta-Analysis basically relies on several measures for integrating research findings including definition of the problem, identifying research studies, coding study attributes, evaluating these findings on a common scale, and examining the aggregation of findings and how they are linked to the characteristics.

When coding the research characteristics, the reported thinking and research is documented in relatively the order. This is primarily because the report incorporates general background or approach, various illustrations of the approach, findings of certain initial research on characteristics, and the outcomes of original research on meta-analytic approach.

Steps in Meta-Analysis and their Application in this Research

As previously mentioned, this research will utilize Glassian Meta-Analysis as the meta-analytic approach in comparing and integrating research findings. This approach consist different steps that are geared towards conducting integrative analysis through statistical methods. While using this meta-analytic approach, the research will incorporate each step in this technique for effective analysis and to generate accurate research findings and conclusions. Some of the major steps that will be utilized in conducting integrative analysis based on Glassian Meta-Analytic approach include:

Formulation of Problem to be addressed

The first step in meta-analysis technique is formulating the issue or problem to be examined, which is relatively similar to any other research work. According to Walsh & Downe (2004), similar to any other research activity, meta-analysis research methodology begins with formulation of the issue to be addressed (p.206).

The problem to be addressed through meta-analysis in a research project is formulated through determining a suitable research question, purpose, and aim. This is the first step in the process because meta-analysis can be regarded as observational evaluation or study of evidence (Egger, Smith & Phillips, 1997, p.1533). This process requires researchers to prepare a detailed research protocol that clearly shows the objectives, hypotheses to be tested, relevant subgroups, and proposed techniques and method for determining relevant studies. During this process, the researcher also writes comprehensive means for collecting and analyzing information from the selected studies.

When formulating the problem to be addressed, the researcher needs to generate a more specific study question, purpose, and aim. This is primarily because broadly framed research questions, purposes, and aims do not necessarily provide accurate research findings or conclusions even if the meta-analysis process is carried out effectively. In this study, the researcher will develop a research proposal that clearly shows the study objectives, hypotheses, relevant subgroups, and proposed research methodologies and techniques. Generally, the proposal will provide a background of the issue and proposed methodologies or techniques for conducting the study. Therefore, the process of formulating the problem or issue to be examined in this research will entail developing a research proposal that contains important information regarding the issue and research process.

The researcher will adopt a more specific approach towards formulating the research problem to be addressed, especially because of the broad nature of the topic or issue of nuclear power. The scope of meta-analysis in this case would be on management strategies for utilizing nuclear power as an energy source and waste disposal in a sustainable way.

The researcher will examine this issue with a view of demonstrating how the use of nuclear power as an energy source can be beneficial in efforts to lessen global warming and climate change. The adoption of a more specific study question, purpose, and aim would help in locating relevant studies that will be utilized in the meta-analytic process. This implies that formulating the research problem or issue to be addressed through a more specific approach helps in easy identification of relevant material for use in the review or analytic procedure.

Locating Relevant Studies

The outcome of the meta-analysis relies and is based on quality and effectiveness of an industrious search of probable studies. This entails determining probable search directions that will govern the robust search of studies on the topic area. Some of these directions that will be utilized to locate relevant studies for the research include using computer search engines, identifying reference lists from studies, electronic databases, and libraries (Shachar, 2008).

The strategy for identifying and selecting appropriate studies for the research process will be extended to include unpublished articles. In essence, the researcher will utilize published and unpublished articles that relate to this subject. The use of unpublished work in this project is fueled by the fact that the findings of this study may differ from published works and restricting the study to published articles may result in distorted findings because of publication bias (Egger, Smith & Phillips, 1997, p.1533).

However, the researcher will define a preferable inclusion criterion through a comprehensive sensitivity analysis to help ensure that both published and unpublished trials utilized for this study are relevant to the topic. The identification of published and unpublished articles will be conducted through electronic databases and manual search of journals. This will also entail using a standardized record form to extract data and avoid errors as well as rating the quality of each study. These measures will help in conducting a detailed scrutiny of each article before determining its use in the research.

The criteria for inclusion of articles in conducting meta-analysis include determining a suitable time period to be covered in the review. In this case, the researcher will examine studies that have been conducted on this topic area from 2000 to date.

The second criterion to be utilized in meta-analysis is a liberal inclusion strategy in which both published and unpublished works will be incorporated.

Third, while the study adopts a more liberal inclusion measure, the researcher will utilize studies that have no severe methodological flaws as a means of ensuring the quality of a study used in this research undertaking.

The fourth criterion is the establishing a control group in which every primary study needs to have a comparison or control group. This is an important criterion in the study since meta-analysis will entail calculating the effect size, which can be described as the mean difference between groups in the form of standard score (Shachar, 2008).

The fifth criterion is ensuring the studies have adequate quantitative data from which the effect size can be calculated. The adequate quantitative data in the studies include sample size, standard deviation, and mean.

Moreover, the researcher will utilize these different measures for deciding which studies to utilize because there are various statistical means for combining the data but no single method is accurate or correct by itself. Synthesizing or combining data for use in this research requires using varying statistical methods in order to enhance the accuracy of research findings and conclusions. Therefore, the researcher will utilize a thorough sensitivity analysis, which is important in evaluating the robustness of combined projections to varying assumptions and criteria for inclusion (Egger, Smith & Phillips, 1997, p.1536).

A thorough sensitivity evaluation will be carried out as part of meta-analysis process because of the divergence of opinion regarding the most appropriate means for conducting a specific meta-analysis. The comprehensive sensitivity analysis helps in examining the robustness of the findings with regards to differing assumptions and opinions. The use of robust qualitative meta-synthesis and thorough sensitivity analysis in this study helps in enhancing the credibility of the research process since these two measures act as quality markers for judging qualitative research. These measures also help in determining studies for inclusion in meta-analysis given that published and unpublished works are used.

The first step towards locating relevant papers or studies to be utilized in this meta-analysis is a robust research that has already been conducted on the topic area or issue. Walsh & Downe (2004) state that it's good practice to conduct a robust search on the topic issue or area in the early stages of meta-analysis similar to the way it's conducted when undertaking a systematic review (p.206). While this process may be relatively inadequate in locating all relevant literature on the issue, it helps in providing direction in the search process and narrowing down to literature that addresses the research issue or problem. Comment by Owner: Since this is chapter three, you should have already located the studies that you want to analyze.

Actually, a robust search on the topic area is relatively a straightforward process in meta-analysis as compared to a systematic review. As previously mentioned, established databases provide a good platform for undertaking a robust search in meta-analysis since they tend to keep studies that are published as journals, produced as books or reported in journal papers. The use of unpublished works in this process is because the depth of meta-analysis synthesis requires the use of such kinds of sources though they are relatively difficult to obtain in some cases.

Once the robust search on studies on the topic area has been carried out, the selected articles will be compiled into an electronic spreadsheet known as Master Database (Shachar, 2008). The compilation of the identified studies enables convenient repetitive sorting and data extraction. This will also help easy selection of the final set of studies that meet the previously mentioned inclusion criteria.

The use of various approaches in deciding which articles to use in meta-analysis is primarily because the qualitative paradigm of this process views truths as multiple unlike classic meta-analytic accounts which only allows the inclusion of controlled, randomized studies (Walsh & Downe, 2004, p.206).

Once the different approaches are utilized in deciding the articles to include in the study, the researcher will synthesize the studies. This process entails translation of one study into another and literally transferring ideas, concepts, and metaphors from one study to another. This implies that the three approaches the researcher will utilize in deciding relevant literature to utilize in this study with regards to meta-analysis include integration of results of several studies in a related field, synthesizing studies from different researchers, and generating quantitative summary of major elements in the various studies or researches.

Collection and Analysis of Relevant Data

Since the problem or issue to be addressed in the research has been formulated and relevant studies identified, the other step in meta-analysis is collecting and analyzing relevant data. This process entails determining an inclusion criterion for selecting articles to be analyzed and ensuring that a suitable approach is utilized for examining the quality of a study (Egger, Smith & Phillips, 1997, p.1533). The researcher is required to consider trials or articles that are relevant to the topic or issue in order to generate the most optimal research findings. The identified studies focus on the problem purpose of this research i.e. effective management strategies for using nuclear energy in a manner that generates social sustainability. The problem purpose of this research provided the basis for searching and identifying studies to be utilized and reviewed. Moreover, the researcher used research propositions or hypotheses as the basis for conducting robust search for studies to be utilized. These research hypotheses or propositions that were utilized in the search are ... Comment by Owner: This should have been done already. So in this section you will need to re write including problem purpose, and propositions (hypotheses)

Hypothesis #1: Public perceptions of nuclear energy are largely driven by disasters.

Hypothesis #2: Public perceptions can influence political decision making.

Hypothesis #3: Unresolved technical issues contribute to policy.

Hypothesis #4: Nuclear energy is not viewed as a viable solution to climate change

The results of each study's rating will be demonstrated in a standardized format to enable comparison of the selected articles. Moreover, the use of these strategies is part of using statistical procedures in meta-analytic review to integrate research findings from various studies and express results through numerical effect-size projections (Bushman & Wells, 2001, p.1123).

The process of analyzing data in this study will also involve calculating the general impact through data combination. However, the researcher will not use a simple arithmetic average to combine data since it would generate misleading results. The methodologies used in meta-analysis to combine data involve weighing average of the results through statistical measures. During this process, the degree of similarity in the studies' findings will be tested for heterogeneity across these studies or articles.

During data extraction and coding, the researcher will review all study in order to identify relevant information and important characteristics, which may linked to the effect size and are linked to the study (Shachar, 2008). The process will entail comparing these studies in terms of their characteristics and findings as well as eliminating any discrepancy. Notably, the data coding process will involve recording research and thinking in the same order based on characteristics described in Glassian Meta-Analysis such as background of the approach and several illustrations of the approach (Glass et. al., 1980, p.1).

Determining Effect Sizes

After data extraction and coding, the next step in this meta-analytic process is identifying the individual and overall effect sizes throughout the studies. In this case, the individual effect sizes across studies should be demonstrated in a standardized manner in order to enable comparison between the identified relevant studies. Once the effect sizes of the individual studies have been obtained, the overall effect sizes should be calculated using statistical measures such as the use of statistical software or computing program. This process will involve using direct weights that are described as the inverse of the difference for every study.

For this study, the researcher will determine the individual and overall effect sizes throughout the relevant studies or researches through the use of previously described statistical methods. Moreover, the researcher will use fixed effects model instead of random effects model for determining the effect sizes across the studies. The use of fixed effects model rather than random effects model is because of the difference in the results and variance in treatment of studies (Shachar, 2008).

Analysis of Bias and Homogeneity

Egger, Smith & Phillips (1997), state that the process of meta-analytic review of studies entails examining homogeneity and bias in studies (p.1533). The analysis of bias and homogeneity is fueled by the fact that the technique primarily focuses on explaining the homogeneity between the findings of individual studies. In this case, the researcher utilizes a more objective appraisal of the evidence in the studies in order to clearly demonstrate homogeneity between study findings and develop accurate research findings and conclusions.

For this study, analyzing bias and homogeneity will entail conducting a synthesis of several studies through appropriate calculations. Moreover, the robustness of research results will be examined based on different assumptions. In this case, the researcher will conduct three analyses of homogeneity and bias to evaluate robustness of research outcomes to different assumptions. The first homogeneity analysis will focus on individual characteristics which will demonstrate chance difference in research findings. Therefore, the researcher will examine whether the differences in the individual characteristics are bigger than those anticipated by chance only (Shachar, 2008).

The second analysis is an analysis of bias in which bias will be detected by examining the left-right symmetry of the plot. Juni et. al. (2001) argues that one of the major concerns in carrying out meta-analysis is the likelihood of bias to emerge when trials with statistically important results are increasingly likely to be published and used. The outcome of the arising bias is that plots of trials' differences or sample size against effect size would always be biased and asymmetrical because of publication bias and other partialities that may affect small trials.

The final analysis of homogeneity and bias evaluation is a Fail-Safe-N evaluation, which focuses on ensuring that significant effects relating to the research issue or problem are identified (Shachar, 2008).

This analysis will be carried out because published article utilized in the research may not publicize significant effects. This analysis will be carried out in light of the unpublished articles utilized in order to demonstrate that the effect sizes would be smaller.

Reporting the Results

The final step in the meta-analysis technique for research is reporting the results once data has been collected and analyzed. In this case, the researcher provides his/her findings from combination of the studies' findings. The researcher not only reports similarities between these articles but also demonstrates the differences in the studies. This is an important part of the process since it demonstrates the applicability of the research findings. According to Bushman & Wells (2001), the findings of a meta-analytic review are usually reported by describing the results through numerical effect-size projections (p.1123).

In this study, the researcher will describe the results of the meta-analytic review or technique through the use of numerical effect-size projections. The projections or estimates will be based on the similarities or differences between the studies that were identified in the analysis process. The researcher will document and describe the findings obtained from combination of the findings from both published and unpublished studies used in this research. The projections are not based on simple arithmetic average of the results but on results generated from statistical measures used in the combination process.

During presentation of research findings, the researcher will demonstrate or report the overall effect size across the various published and unpublished studies utilized in this study. Shachar (2008) states that an overall effect size that is calculated from a huge sample is increasingly likely to be accurate as compared to an overall effect size calculated from a small sample. Therefore, using published and unpublished works implies that the study will have a huge sample through which increased accuracy of results can be generated as compared to using a small sample.

The researcher will also utilize graphical display of results because the graphical display of findings from individual studies on a standardized scale is a vital step that enables a visual evaluation of the extent of heterogeneity between various studies. Egger, Smith & Phillips (1997) state that results or findings from every trial are beneficially graphically displayed alongside their confidence intervals (p.1535).

With regards to confidence intervals, the margin of error in meta-analysis will be quantified or calculated through the concept of a 95% confidence interval. According to Egger, Smith & Phillips (1997), meta-analysis results are better reported and understood when they are graphically displayed while their effect sizes with 95% confidence intervals are displayed through the use of a Forest Plot or through a histogram of the distribution of the effect size (p.1535). In this case, the researcher will choose the most appropriate means for graphically displaying the effect sizes with 95% confidence intervals.

The use of confidence intervals is attributed to the fact that whenever a parameter is estimated in meta-analysis of studies, the distribution of the specific estimator needs to be determined. The researcher not only provides a point estimate or projection of the parameter but also obtains a confidence interval (Shachar, 2008). If the process for computing or calculating a 95% confidence interval is used several times, the interval will have the actual parameter value 95% of the time. In this study, the parameter of interest under examination is the individual and overall effect sizes across the studies.

Qualitative Interpretation of Effect Size

The final step in meta-analysis in this study is conducting qualitative interpretation of effect size i.e. the individual and overall effect sizes across studies. This is an important step because interpretation of results or outcomes in meta-analysis technique requires an understanding of standards utilized in order to promote meaningful and beneficial interpretation of individual and overall effect sizes. The existing measures for qualitative interpretation of effect size not only evaluate the meaningfulness of an effect size but also provide guidelines for conventional measures for this process (Shachar, 2008).

In this study, the researcher will conduct the qualitative interpretation of effect size through using the method suggested by Cohen as part of statistical power evaluation for behavioral sciences. The researcher will qualitatively interpret effect size based on 0.2, 0.5, and 0.8 as minimal, moderate, and meaningful effect respectively (Cohen, 1988). Through this method, the researcher will classify effect sizes into three major categories or groups i.e. small, medium, and large. The small effect size category will be for effect size below 0.32 while medium and large will be for effect sizes between 0.33 and 0.55 and greater than 0.56 respectively (Lipsey, 1990).

Expert knowledge and thorough research are critical because of the inherent need for rigor in qualitative research, the field having essentially shifted away from quantitative concepts such as reliability and validity. The concept of rigor should include thoroughness in research, the development of expert-level knowledge of a subject, and the application of unbiased analysis of the information that is available (Morse et al., 2002). The absence of any of these things in the Meta-Analysis Technique can undermine any research, including quantitative, but in qualitative research the researcher must pay extra care and attention to issues of bias and completeness, not passing up any potentially valuable information.

One of the significant issues with the Meta-Analysis technique is that there will inevitably be gaps in the existing knowledge. The purpose of the research is therefore to piece together what truths are known, and identify the different areas where future research would benefit. Ultimately, the methodology has to move beyond the literature review and arrive at generalizations that can be more broadly applied, also known as empirical generalization.

There are a number of approaches to generalization. First, the researcher must answer the question as to whether the material uncovered can be generalized. Second, the researcher needs to examine this with respect to each different study. Third, the researcher has to understand whether theories can be derived from the research in question (). In this case, no theories are being sought.

However, the first two questions about generalization are valid for this research. It is reasonable that no one study can be extrapolated globally, because of the role that cultural context plays in the perceptions and applications of nuclear power in each national context, but it is also reasonable that by examining findings on these questions from a wide number of countries, that any commonalities that span enough countries, and with enough strength, could be generalized and those findings apply to any similar situation.

Of course, there will be limits in this. For example, findings from Anglo-tradition democracies may be extrapolated to Germanic or Franco-tradition democracies, but the research has noted that this is not necessarily the case. There is less applicability of such findings to the Japanese context given the cultural gap between Anglo-tradition democracies and Japan. And there might not be any applicability to Russia, Iran or other country that essentially has no tradition of democracy and whose cultural roots are dramatically different from those in Anglo-traditional countries.

The optimal technique for the research questions at hand is known as research synthesis, which is the qualitative equivalent of meta-analysis (Jefferies, 1999). Research synthesis is a technique by which the researcher examines the sum total of the literature on a given subject matter in order to identify trends, as well as gaps in the literature. A research synthesis is thus a form of evolved literature review, wherein the researcher subjects the research to logical and systematic patterns of evaluation (Jefferies, 1999. The research synthesis will be based on published and non-published studies that the researcher will identify and select for this research. The number of studies to be utilized in this study will be described towards the end of this chapter.

The research synthesis is optimal for the social sciences, because much of the existing body of research is qualitative in nature. It is also optimal in situations where the existing body of research on a subject is somewhat thin, with gaps. Where research on a subject is specific, dense and quantitative, a meta-analysis would be more appropriate. The body of existing literature on the subject of this report does not meet the criteria for a meta-analysis, being understudied, disparate and qualitative in nature. Thus, the most appropriate research methodology is the research synthesis. Unlike literature review, research synthesis is a quantitative research methodology that is used for objective and effective review of completed research or studies (Jefferies, 1999).

The other aspect of research tradition is multi-disciplinary study. While multi-disciplinary study does not always lend itself to the narrow focus of academic journals, the real world is often multidisciplinary. Actually, multidisciplinary study is increasingly being recognized as a necessary component for answering complex questions, resolving multifaceted problems, and obtaining logical understanding of complex issues that cannot be resolved sufficiently or addressed comprehensively by any single discipline ("Defining Interdisciplinary Studies," n.d.). There is a time and place for laboratory questions that exclude real world factors in order to isolate variables, but this is not the time.

Research about nuclear power and its role in our energy future is by its nature multi-disciplinary as it lies at the intersection of nuclear science, sociology, psychology and political science. Journals from all of these disciplines have contributed to this study. There is not just one variable, acting independently, that governs this issue. It is indeed many variables, and they sometimes work together to influence opinions and actions (Bauer, Brecha & Luderer, 2012, p.16807).

The outcome of meta-analysis, across disciplines, in a qualitative research format, is going to be research that is more accurate and reflective of the real world problem that is to be solved than would be possible in a controlled experiment. The fact that the data needed does not always exist, may not be accessed, or that there are ethical issues with conducting experiments in this field only add to the necessity to conduct qualitative research in this area. Walsh & Downe (2005) state that meta-analysis is an important method for qualitative research since it helps deepen understanding of contextual dimensions of the issue being examined (p.204).

Studies to be researched

The researcher has identified 20 studies in the meta-analysis on this research issue. These studies will comprise published and unpublished articles addressing the issue of nuclear power, especially with regards to reducing climate change. The major criterion for the identification and selection of these studies will be their relevance to the research issue, particularly on how nuclear energy may be the most suitable solution to climate change. In addition, the researcher will employ other criteria for selecting the studies since the research issue is broad and has attracted considerable attention in the recent past. Comment by Owner: Has identified.

As previously mentioned, the studies to be researched will be both published and unpublished articles relating to the topic area. The selection of the studies to be researched will be based on the liberal inclusion criteria that were previously described. Apart from using published and unpublished articles, the inclusion criteria will include limiting the robust search of the topic area to studies conducted between the years 2000 to date (Shachar, 2008). With regards to the quality of a study, the researcher will only utilize studies that do not show severe methodological flaws. Moreover, these studies must have adequate quantitative data and a comparison or control group that can be utilized in calculating the individual and overall effect sizes.

One of the key measures in determining studies to be researched is their relevance to the research topic or area. This is a major component in selecting studies to be researched because the research area is a broad topic that has attracted many studies because of the controversial nature of nuclear power in relation to its use as an energy source, environmental harms, and benefits. The 20 studies to be reviewed were identified through the robust search of the topic area, which is considered as an important process when conducting meta-analysis as previously mentioned. Comment by Owner:

While robust search of the topic area will be used as the first means to identify studies to be researched, the researcher will adopt a more specific approach in order to ensure relevance of these studies to the research topic. The researcher will have preference of journal articles over other types of articles because of the likelihood of published and unpublished articles to meet all requirements in the inclusion criteria. The preference of journal articles over other types of articles is also fueled by the fact that they seem to be credible as compared to other kinds of articles.

Akobeng (2005) claims that meta-analysis primarily relies on the validity or credibility of the studies (p.846). As a result, a researcher is required to explicitly show the kinds of studies included in the meta-analytic review based on the suitability of the articles on the study design and research question, purpose, and aim. In this study, the researcher will utilize different electronic databases since a single electronic database search lacks sensitivity and may make the researcher miss relevant articles.

The researcher will utilize different or multiple sources to conduct comprehensive search and avoid the issue of publication bias. A comprehensive search will be utilized to identify studies to be researched since extensive searches help to avoid publication bias that may occur when trials with statistically important outcomes or findings and increasingly likely to be published and cited, particularly in English language journal articles.

When conducting the searches, the researcher will utilize the established research question to guide the search and help ensure that relevant articles are identified. In this case, the researcher will use keywords and synonyms from the research question to conduct the extensive search in various databases with the aim of identifying articles for inclusion in meta-analytic review or study (Akobeng, 2005, p.846).

Research Questions and Hypotheses

This research is by its nature primarily qualitative. However, the researcher will also utilize quantitative methods, especially to develop research questions and hypotheses. As previously mentioned, this research will involve the use of both qualitative and quantitative research methods though much focus will be on qualitative methods. As a result of the use of quantitative studies, there researcher will test four hypotheses in this study.

The first hypothesis reflects on the effect that disasters have on public perceptions of nuclear power. For obvious technical and ethical reasons, it is not possible to induce a nuclear disaster in order to study this issue knowledge must be gained from the studies that have already been conducted. There have been three high profile nuclear power disasters at Three Mile Island, at Chernobyl and at Fukushima.

There is information available from surveys that were conducted in the wake of Chernobyl, and there have been numerous studies in the past decade about nuclear power acceptance by the public. These recent studies include some specific before-and-after data about Fukushima. The researcher will review at least four studies from surveys that were conducted in the aftermath of Three Mile Island and Fukushima incidents. In essence, the researcher will review two studies conducted in the wake of Chernobyl and two studies conducted in the aftermath of Fukushima incident.

To address the first question (Hypothesis #1:Public perceptions of nuclear energy are largely driven by disasters), then, it will be necessary to review the literature, in the style of a meta-analysis, in order to determine whether or not public perceptions have changed before and after the Fukushima meltdown. This information can then also be compared against the information that exists from Three Mile Island and from Chernobyl.

One of the most valid reasons for conducting a meta-analysis on nuclear energy, specifically the fixed effects model is that this type of research can allow for more complex concepts to be visited than would be possible via other means. This topic has four distinct hypotheses, and they are multi-disciplinary in nature. Moreover there are specific technical and ethical barriers to answering these questions in a laboratory-style study design. Even today, it would be impossible to effectively gauge public reactions to Fukushima, as that was four years ago.

On the second question, (Hypothesis #2: Public perceptions can influence political decision making) there has been a lot of data gathered about this. Much of this data, unfortunately, is not in the public domain. Political parties would be the ideal source of this information, as they know exactly what drives people to vote, and can probably express this as well as anybody. The problem, of course, is that this information is gathered for proprietary purposes, and ultimately this data is unlikely to be made available to the public while it still has value, as this is a key source of competitive advantage for political parties.

The academic information on this is somewhat limited. To study this would entail staging an election instead there are proxies used. One proxy is to substitute voter action for voter intent, though in the real world those two things can diverge significantly. While experts have been able to predict voter intentions quite well (Singh et al., 1995), nuclear power is fairly low on the list of issue that people take into consideration when voting. It may be entirely impossible to predict voting behavior by comparing voter's beliefs about nuclear power and a politician's stated policies on the matter.

While such information would clearly answer the question at hand: (Hypothesis #3: Unresolved technical issues contribute to policy): it has never been gathered directly, and the reality is that there is going to be a lot of noise in such a survey. Again, where hampered by a lack of data, it is necessary to rely on qualitative techniques such as meta-analysis to provide the rich level of description required to understand this complex, multidisciplinary subject.