Introduction
The purpose of this study is to determine the association between asthma and smoking status among adult African immigrants in California. I intend to use a quantitative correlation approach to explore the association between asthma and smoking status and investigate the influence of selected demographic variables (age, sex, years since immigration, marital status, alcohol use, education level, income level, and employment status) on this same relationship.
The independent variable will be smoking status, and the dependent variable will be asthma. The hypothesis states that there is an association between asthma and smoking status among African immigrants living in California. The findings of this proposed research could help to fill a research gap, which exists because previous researchers have not extensively explored the relationship between the aforementioned variables among African immigrants in California.
This chapter contains five major sections. The first part is the research design and rationale section. It explains the research variables and their connection to the research design. The second section contains details of the methodology used in the research. In this section, the information about the target population, sampling procedures, power analysis, instrumentation, operationalization of constructs, data management, data analysis, procedures for recruitment, participation, and data collection are outlined. The third section of this chapter outlines the threats to validity, while the fourth section outlines the ethical procedures governing the research process. The last section of this chapter is a summary of the main tenets of the chapter.
Research Design and Rationale
As mentioned in other sections of this research, the purpose of this paper is to determine the association between asthma and smoking status among adult African immigrants in California. The dependent variable is asthma status, while the independent variable is smoking status. The moderating variables are age, sex, years since immigration, marital status, alcohol use, education level, income level, and employment status.
The research design for this study is the quantitative correlation approach. This design is often used to explore the association between two or more variables in research (Armijo-Olivo, Stiles, Hagen, Biondo, & Cummings, 2012). There are two variables in this study, as seen from the research question, which states, “What is the association between asthma and smoking status among adult African immigrants in California?” There is an association between the design and this research question because the variables stated in the latter are quantitative (measurable). For example, asthma status is often measured in quantitative metrics, and smoking status is also measured quantitatively.
Furthermore, the California Health Interview Survey (CHIS) dataset used in this study measures the two variables in numbers (quantitative). These factors limit the application of the qualitative technique, which measures subjective outcomes. Collectively, these factors explain the justification for using a quantitative approach. The use of the correlation design is also appropriate for this study because it measures two or more variables (Chan et al., 2013).
In this study, I seek to explore two variables, which are smoking and asthma status. Indeed, the research design has a connection to the research question. The use of the quantitative correlation approach comes with limited time and resource constraints because the relationship under investigation (the association between asthma and smoking status) usually involves a deep analysis of data and a careful evaluation of demographic data to understand the relationship (Hassan et al., 2015).
However, the use of secondary research data to investigate the same relationship has alleviated this problem, and the time taken to explore the association between smoking and asthma is small. Furthermore, since secondary data is used in this research, there are limited resources needed to find out the association between the dependent and independent variables because the data is already published and does not require the researcher to use additional resources to get it.
The quantitative correlation design chosen for this study is consistent with other research designs capable of advancing knowledge in the health care practice because it opens up abundant opportunities for understanding the association between asthma and smoking status for future researchers who may want to explore the association further (Hoffmann, Bennett, & Del Mar 2013). However, it is important to point out that the research design is a form of descriptive approach and does not necessarily explain which of the variables studied influences the other. In this regard, it provides a good starting point for investigating the relationship between asthma and smoking by allowing researchers to understand the strength and direction of the association between the two variables without necessarily explaining the details surrounding the causation. Thus, future researchers can narrow the findings down to understand the intricate details surrounding the causation, experimentally, or otherwise.
Methodology of Research
Target Population
In the context of this study, the target population refers to the selected immigrant group, which I am studying. As highlighted in the first chapter of this paper, there are inadequate research studies that have investigated the association between asthma and smoking among adult African immigrants. Based on this gap, the target population for this research is adult African immigrants in California. This target population comprises of men and women above 18 years of age. Additionally, they must have originally emigrated from Africa and assume the status of first-generation immigrants living and possibly working within California State, USA.
The target population size is 1,946 people. This number is expected to include the entire set of respondents for which the research data would be based on. I estimate that this sample size is large enough to make inferences about the association between asthma and smoking status among adult African immigrants in California because the CHIS database only includes about 21,000 respondents, while Africans comprise a small share of this group (UCLA, 2017). Thus, using a sample of 20,000 people, a sample population of 1,946 is estimated to be around 9% of the total population sample in the CHIS database. This percentage is within the sample size requirements outlined in most social research studies needed to make inferences about statistically significant relationships (Faber & Fonseca, 2014).
Sampling and Sampling Procedures used to Collect Data
As highlighted in other sections of this report, the main sources of data and information for this study would be the CHIS database. Since my main sources of information are attributed to these three sources of secondary data, the sampling strategy mirrors the same plan present in the secondary data. There are four main types of sampling strategies – random sampling, stratified random sampling, systematic sampling, and rational sub-grouping (Mathie, Roniger, & Van Wassenhoven, 2012). The CHIS dataset was developed through a random sampling technique where respondents had an equal chance of being selected for the study (UCLA, 2017).
This sampling strategy is often lauded for reducing selection bias and improving the reliability of the associated findings. Additionally, the CHIS dataset was developed by randomly selecting one adult respondent in each household that chose to participate in the study. The sampling strategy employed in the study was designed to meet two main goals. The first one was to provide local estimates of population-based health data for comparison across different counties in California, while the second one was to provide statewide population-based health data for all the ethnic and racial groups in California (UCLA, 2017).
The sampling strategy adopted in the CHIS dataset was drawn from the dual-frame random-digit-dial (RDD) method, which included a sample of the respondents’ views using telephone surveys (UCLA, 2017). Besides the landline sample, the CHIS dataset also included a statewide cell phone sample of the overall population. The landline and cell phone samples described two separate groups of respondents that characterized the data.
The researchers administered these samples through a computer-assisted telephone interview that included both the statewide landline random digital dial and the statewide cell phone sample. The landline sample was stratified according to county demarcations, groups of small counties, and sub-county areas (UCLA, 2017). Based on the nature of the data collection method, only those households that had a landline telephone were included in this sample.
The sampling frame used to develop the CHIS dataset involved the use of traditional random digit dial and cell phone random digit dial sampling frames. The inclusion criterion was households, which had a landline or cell phone. The inclusion criterion also included the views of respondents who were adults (18 years and above) (UCLA, 2017). Comparatively, the exclusion criterion included respondents who were under 18 years and households that lacked either a cell phone or a landline. These frames were used to get the views of different respondents in the state of California by understanding the source materials and devices from which the sampling population was drawn from.
Power Analysis
Power analysis is an important aspect of this methodology because it outlines the procedures used to select the right sample size that could be reliably used to find out if there is an association between asthma and smoking status among adult African immigrants in California. To determine the sample size, I used a reliable calculator developed by Creative Research Systems. Using this tool, I obtained a sample size of 1,946 based on a confidence level of 95% and a confidence interval of 0.5. This confidence level represents the level of surety that the inferences I will make after assessing the available data are reliable. The confidence level also insinuates the same position because it outlines the degree to which I am confident that my findings are true and reliable.
I used a confidence level of 95% because I wanted to have a reliable sample. Indeed, a confidence level of less than 1.0 denotes a high level of precision for the sample used. Similarly, a confidence level of more than 1.0 symbolizes a lower degree of precision. I assume that the effect size of 1,946 people is adequate for drawing reliable conclusions about the relationship between the dependent and independent variables for the target population – African immigrants. The confidence level of 95% was used because many researchers who have done similar studies use it as a standard measure of confidence for their findings (UCLA, 2017).
Procedures for Recruitment, Participation and Data Collection
This study is a secondary analysis of archived data (CHIS secondary dataset). CHIS researchers originally collected this data. As mentioned in this study, this database is the largest in California. The main variables analyzed from the database include asthma incidences and smoking incidences. The analysis will include an attempt to statistically interpret the relationship between both variables. Some researchers, such as Faber and Fonseca (2014), who say that it is quickly becoming a popular method of increasing the efficiency of health care research have supported it. The benefits of using this data for this type of analysis are clear, but the barriers are also many, as argued by Faber and Fonseca (2014) who say the success of this type of data collection depends on the efficiency of health agencies and researchers to conduct reliable public health studies.
Data Management
Explaining the procedures for data management is important in this research study because it affects the credibility of research information (Jain et al., 2015). Data management also involves explaining the procedures for accessing data and the necessary permissions required to obtain such information. In the context of this study, the data analysis process explains the reputability of the information retrieved from the CHIS, contains an explanation of why the information retrieved from this secondary data source is important and represents the best source of data for the research. I gained access to the database through an open access online platform available at CHIS. Therefore, there was no need for getting special permission to access the articles.
Although it is common practice that researchers get special permission to use secondary research data (Jirojwong, Johnson, & Welch, 2014), this requirement is mostly applied to research articles that are not freely available online. However, this requirement was not applicable in my review because the information used is publicly available. However, I will seek permission before analyzing these CHIS secondary data.
Although using freely accessible data is acceptable in this study, it is important to use reputable data sources when undertaking this type of research. In the context of this study, the reputation of the findings, which I will produce, will only be as good as the reputation of the secondary data used. As mentioned in this paper, the main data source was the CHIS dataset. This dataset was appropriate because it is statewide and is undertaken by a reputable organization. Furthermore, UCLA (2017), which undertook the study, has been collecting similar data since 2010 with relative success. Based on its good reputation, different professionals, including journalists, policymakers, and health experts, use its information. Thus, this source of data is appropriate for this study.
Instrumentation and Operationalization of Constructs
Instrumentation of Constructs
As mentioned in this study, the main source of data is CHIS. It is among the largest health interview survey in America and is conducted on an annual basis to provide information about different health topics (UCLA, 2017). Although the CHIS publish health data relating to different years, the data used in this report was published in 2015.
Looking at the appropriateness of the CHIS data to the current study, I find that it is relevant and specific to the topic under investigation because the dataset provides health data relating to different ethnic and racial groups in California. A focus on African immigrants as one cohort in the study is ideal for my analysis because the current study focuses on this ethnic group as the target population. The dataset is also appropriate to the current study because asthma is a health topic investigated in the research data. Other health issues highlighted in the dataset include diabetes and obesity (UCLA, 2017). The inclusion of asthma as a relevant health issue in the dataset and the provision of health data relating to immigrants bring my attention to the appropriateness of the data to the current study.
As highlighted in this chapter, the findings contained in the CHIS document are freely available to the public. Therefore, there is no special permission from the researcher needed to use the instruments. Based on the availability of the health data outlined by CHIS, different people, including policymakers, state agencies, and community organizations, find the resource useful in improving the health outcomes of their subjects.
A review of the reliability and validity of the values outlined in the research study revealed that both metrics are desirable because the CHIS data acts as a model for collecting state and local health data. This attribute demonstrates its reliability. Its use in advanced sampling and application methodologies also adds to the same metric because such evidence has been used to influence key policy changes in different areas of research (Mertz et al., 2014). Its credibility is also supported by the fact that the database contains information that would appeal to key stakeholders in the health sector. These published reliability and validity values show that the sources of data are relevant to the current research.
As highlighted in this study, the CHIS database was used to collect health data across different ethnic and racial groups in California. In the past, it has been used to collect health data from all 58 counties in the state of California (UCLA, 2017). However, there are cases where the researchers have oversampled specific areas within the state that are heavily populated (such as Los Angeles and San Diego).
The reliability and validity of the findings developed from the past use of CHIS have been confirmed through the involvement of large and diverse samples (UCLA, 2017). In other words, past users of the data established that the samples used in the dataset were representative of the ethnic and racial diversity of the state, particularly because the findings could be used to answer specific and important health questions about different ethnic and racial groups in the state.
The CHIS data has sufficient instrumentation to answer the research question because of some sampling and methodological issues highlighted in this chapter. For example, I have established that the dataset contains health data about different ethnic and racial groups in California. This one instrumentation is useful in answering my research question, which focuses on one ethnic group – African immigrants.
Other aspects of the instrumentation used to develop the CHIS are its consistency, flexibility, and adaptability. These attributes mean that the findings included in the report can be used to investigate new health issues and emerging trends among specific racial or ethnic groups. Lastly, the inclusion of spatial and geographic data in the CHIS dataset is another aspect of its instrumentation that would help to answer the research question because I can narrow down to a specific locality (California), which is at the core of my analysis. This type of instrumentation is often unavailable in national health data (Liamputtong, 2013).
Operationalization of Constructs
The research variables identified in the CHIS database were 45. In this study, two variables were relied on to complete the research. They include the independent variable, which is smoking and the dependent variable – asthma. In the CHIS database, race emerged as one variable that helped to define the target population – African immigrants. It was denoted by the term RACE CN-P1. Seven values were attached to this variable. They appear below.
Smoking is also another variable that was included in the CHIS database and it was denoted by the code AE15A. This variable indicated the number of people who were currently smoking, had smoked before, we’re not sure whether they were smoking, or had smoked before. These statuses were represented by unique codes that ranged from 0 – 9. The same was true for asthma as a dependent variable because it was denoted by code AB 41.
This variable helped to ascertain the number of people who had suffered from asthma incidences, or not. It was also used to affirm the frequency of asthma attacks among the target population. To collect the right data for analysis, I only investigated the relationship between the independent and dependent variables within the African population from the CHIS dataset, which was denoted by code 4 mentioned in the table above. The number of respondents was 1,946.
Dependent variable
As mentioned in other sections of this paper, the main variables for this paper are asthma status (dependent variable) and smoking status (independent variable). These variables are operationalized depending on how well they are categorized into measurable parts. The process of measuring the variables will be done empirically and quantitatively. The operationalization of asthma status as the dependent variable lies in my understanding of the definition of the health condition, which manifests as a long-term inflammatory disease of the respiratory tract (Tamimi, Serdarevic, & Hanania, 2012).
These symptoms make it difficult for their victims to breathe. In the context of my analysis, the dependent variable would be operationalized by noting the changes in the number of people who present the symptoms of the disease. These symptoms would have to be clinically proven to reach the threshold for diagnosing a patient as suffering from the condition. Changes in asthma incidences within the sampled population would also be measured as an indicator of the number of people who manifest long-term symptoms of the condition and not necessarily those who manifest the same symptoms in the short-term. Thus, the population determined to be falling under this health group must have manifested long-term symptoms of the condition.
The variable would be measured or manipulated by observing changes in the percentage of African immigrants who have asthma and are smoking. Efforts would be made to observe whether there are noticeable patterns between the two variables. The variable/score would be calculated by analyzing changes in asthma cases among people who smoke and those who do not smoke. This score would be represented as a percentage because I will describe the percentage of people who smoke, viz-a-viz those who do not smoke. In this regard, there would be two scores. One of them would be the percentage of African immigrants who have asthma and smoke, while the other will be a percentage of African immigrants who have asthma and do not smoke.
Independent variable
Comparatively, the independent variable will be smoking status. This variable will be operationalized by noting that the associated changes are reflective of the number of people who inhale tobacco smoke. Therefore, the measure does not include people who smoke other substances, such as marijuana, or those who partake in other types of smoking. This variable is also operationalized based on the frequency of smoking because there is no criterion for categorizing people as smokers, based on the times or frequency they engage in the habit. For example, a person who has smoked for only one week and another who has smoked for several years are included as equals when operationalizing this variable.
However, the age criterion for including the same respondents is valid because, in this study, I will only include adult smokers and not teenage smokers. In other words, this variable will only include people who are smoking and above 18 years. Similarly, this analysis does not have a gender criterion because both male and female smokers are included in the assessment. At the same time, changes in smoking status are also observed in people who have quit smoking and are currently smoking. Therefore, people who have smoked at any point in their lifetime are included as part of the review.
The variable would be measured by counting the number of people who smoke or who have smoked in the past. This measurement would be a numeric figure. The scale score for this variable will be calculated based on the number of people who smoke, compared to the population of African immigrants who do not smoke. In this regard, the computation would assume the form of a fraction, where the numerator would be the number of people who smoke and the denominator would be the number of people who do not smoke.
Data Analysis Plan
The data analysis plan is an important part of this research process because it would help me get useful information/findings from the research inputs. To undertake this process, I will use the SPSS statistical software tool. The 2015 version will apply to the study because it contains the latest updates and tools for data analysis (Arkkelin, 2014).
Data Cleaning and Preparation
Before using the aforementioned SPSS software tool, I will participate in data cleaning and screening procedures to make sure that the inputs of the data analysis process are reliable and credible. However, I do not expect to have many errors or mistakes associated with the data inputs (used in the data analysis process) because the dataset used is from the CHIS, which is often a reliable and valid source of information.
Nonetheless, to make sure that the information keyed in the SPSS software is credible, I will look out for the range data. This process will also involve checking for the minimum and maximum values associated with the variables of interest by analyzing scores associated with the descriptive data (Murari, 2013). To make sure that the data is clean, I will also look out for “abnormal” responses. This process will help me understand whether the responses included in the analysis are legitimate, or not. My criterion for doing so would be counterchecking suspect responses with those of the majority respondents.
The data screening process will also involve looking out for identical responses from the data to avoid cases of duplication (Perry, Barak, Neumann, & Levy, 2014). To make sure that this process is effective, I will investigate if there are cases of identical values, or near-identical values, to establish whether they have been posted erroneously, or they represent the same scores. If such cases are confirmed, I will choose to maintain one case/value and remove the duplicate value. The last step in the data screening process will involve manually checking for errors and mistakes associated with the data. Unlike other data screening processes, there would be no specific analysis aimed at detecting specific issues because the focus would be to investigate whether there are any cases of oddities (Shaughnessy, Zechmeister, & Zechmeister, 2014).
One criterion I will be using is looking out for “empty cases” where the values involved would be missing or insignificant. When such cases are established, such data would be omitted from the data analysis process.
Statistical Analyses Plan
Descriptive Statistical Analysis
The dependent and independent variables would be described using descriptive statistical methods that include frequency tables and measures of central tendency. The frequency statistics would simply help to point out the number of times each variable occurs. For example, it is instrumental in explaining the frequency of asthma among people who smoke. The measures of central tendency would also be useful in describing the data by using one unit that provides a general description of the scores. This statistical tool would be useful in describing the personal information of the sample. For example, researchers have used it to provide details about gender, educational qualifications, income, and such as demographic data. This type of information is applicable in understanding some of the moderating variables such as age, sex, years since immigration, marital status, alcohol use, education level, income level, and employment status that are highlighted in this study.
Inferential Statistical Analysis
Bivariate Analysis
As part of the data analysis plan, the use of inferential tests will be central to the whole process because it would help to test the hypothesis, which states, “There is an association between asthma and smoking status among adult African immigrants in California.” The bivariate analysis method is applicable in this study because there are two main research variables – asthma and smoking incidences. The bivariate analysis would be useful in understanding whether there is a relationship between the two sets of values (Polit & Hungler, 2013). There are different types of bivariate analysis methods, which include scattering plots, regression analysis, and correlation co-efficient (Polit & Hungler, 2013).
The relationship between the independent and dependent variables in this paper will be analyzed using a scatter plot diagram. This technique would help to provide a visual idea of the patterns that would emerge in the research to possibly explain a relationship between both variables. This scatter plot will be developed from the SPSS software package under the “graphs” tab.
Multivariate Analysis
Unlike the bivariate technique that analyzes two variables at the same time, the multivariate analysis analyzes the relationship between multiple variables. This type of analysis will help us to mimic the real-life scenario of factors that affect the relationship between the dependent and independent variables. This analysis will be crucial in investigating the effect of mediating factors such as age, sex, years since immigration, marital status, alcohol use, education level, income level, and employment status on the dependent and independent variables. The one-way MANOVA technique would be the main statistical test used in the study. Its significance would be hinged on investigating whether there is a significant difference between two or more groups of variables.
Data Analysis Matrix Table
The data matrix table in appendix 1 provides a visual summary of the descriptive analysis in the study. In the matrix table, there are four main concepts, which are aimed at describing the smoking behaviors among the immigrant population, understanding the occurrence of asthma within this population group, ascertaining the number of African immigrants highlighted in the study and understanding how their demographic profiles affect the relationship between the dependent and independent variables. There is a common data source that would be used to collect this data – California Health Interview Survey (CHIS) 2015 Adult Data.
Each of the concepts described in the data analysis matrix has a unique level of measurement, which is either ordinal or nominal. I allocated ordinal measurements to concepts that could be described in the form of a range of quantitative variables. For example, the incidence of smoking and asthma has described this way because they included a range of statistical numbers. The same analysis is true for the number of African immigrant population in the study because the variable is an estimate of the total number of African immigrants living in California. Lastly, the description of the demographic profile of the target population was measured nominally because they mostly included variables that were in one or two states. For example, gender could only be measured in terms of male or female states.
Educational qualification could be measured using a continuum of different states, such as high school diplomas, undergraduate education, postgraduate education, and the likes. The use of different levels of measurement in the analysis justified the use of the ordinal and nominal measurement techniques for describing the demographic profiles of the respondents. The data analysis techniques used in the descriptive analysis were either frequency tables or measures of central tendency.
Mock Tables
The mock tables highlighted in appendix 2 paint a picture of how the statistical reports would be analyzed. The tables largely represent the findings of the descriptive and inferential analyzes and are tabulated using percentages and relevant statistical analysis tools. It is important to point out that each table is representative of a research study objective and has a detailed analysis of the procedures used to come up with the data.
Although I will use multiple statistical tests for this review, I will account for the possible errors that may occur from doing so by putting a stricter significance threshold for each test. For example, to affirm a positive association between asthma and smoking status, I will only consider values that are 0.5 to 1.0. This range is stricter than the normal range of proving such a relationship, which often varies from 0.0 to 1 (Wright-St Clair, Reid, Shaw, & Ramsbotham, 2014).
Part of the data analysis process will also involve the inclusion of mediating factors, such as sex, income, and education (among others), in the analysis. The justification for including these variables is their significant influence on smoking behavior and asthma development among the target population. Furthermore, it is difficult to have a perfect experiment because the compliance rate will not always be absolute and neither is the dropout rate. Consequently, introducing covariates in the analysis will help to clarify the relationship between the dependent and independent variables (Arkkelin, 2014).
As highlighted in this chapter, the data analysis process will involve inferential analyses using correlation tests. Testing for correlations will involve analyzing values that range from -1 to 1. If I get values that are closer to 1, I will deduce that there is a strong relationship between smoking and asthma status among African immigrants in California (Rezaei et al., 2017). The opposite is also true because if I deduce a value that is closer to -1, I will assume that the two variables share a weak relationship. A value that is closer to 0, or 0 would mean that there is minimal or no linear relationship between the two variables as well (Mohini & Prajakt, 2012).
Threats to Validity
Threats to validity are issues that could affect the credibility of the research findings. Indeed, as Nieswiadomy (2012) observers, threats to validity may significantly undermine the reliability of any given research design. The main threat to validity for the current study is centered on the use of the CHIS database, which was the main source of information for this study. The failure to be part of the original research is one threat that suffices in this context because it means I was not privy to pertinent research information that could be instrumental to the current research. One threat to the external validity of the research is that it could not apply to migrant populations outside the context of the research region – California.
Furthermore, it could be inapplicable to other racial or immigrant groups living in California because the current research only focuses on African immigrants in the state. The specificity of variables is also another threat to the internal validity of the study because different criteria for analyzing the variables may cause distortions in the findings. For example, a person’s smoking status may quickly change within the time a study is undertaken. This change may significantly affect the outcomes of the study.
To overcome threats to internal validity, I will obtain as much information about my sample population (African immigrants in California) as possible to get a better understanding of the respondents and to evaluate how their social, political, and environmental dynamics affect the study. This measure is useful in solving the internal threat associated with instrumentation (Lee, Crawford, & Wallerstedt, 2012). Standardization is also another technique I will use to overcome the same threat. In this approach, I will strive to consider the conditions under which the original study was carried out when developing the research findings because I am depending on the CHIS methods and data for my analysis.
I could also experience some of the same validity problems when analyzing external validity issues associated with the study because there is a problem associated with applying the findings of this study beyond the selected sample. One problem I could experience from this issue is the possibility that the secondary data used to develop the research findings may not apply to a different time, outside of the publication date (Fernandez-Hermida, Calafat, Becoña, Tsertsvadze, & Foxcroft, 2012). The secondary data used in this study was published in 2015, and this means that the findings may not apply to any other period under review.
I may also experience the same problem with instrumentation because I have already established that the secondary data used was developed from telephone and cell phone surveys. The use of different instrumentation techniques in future research may yield different results. For example, if researchers used one-on-one interviews to collect data, as opposed to the telephone and cell phone surveys, the relationship between asthma and smoking status may change because the context of the investigation would change, possibly in favor of respondents feeling more obligated to take the research more seriously, as opposed to a phone interview (Arkkelin, 2014).
To overcome the external validity issues mentioned above, I will point out, in the findings section, that the results apply to a specific period. This way, the users would know the limitations of the study and consider the same when making future or past inferences about the relationship between asthma and smoking status among African immigrants living in California (Lang & Altman, 2014). This solution would be instrumental in solving external validity issues associated with the publication date for the CHIS data. To solve the problem of instrumentation, it is essential to include co-variation in the study when analyzing the data to understand the effects of the use of different instrumentation techniques when developing the findings (Lang & Altman, 2014). Similarly, standardizing the research process could help to avoid this issue because it would make sure researchers replicate the same context when conducting future studies (Lang & Altman, 2014).
Lastly, I do not anticipate any threats to construct or statistical conclusion validity because I have already explained how each variable will be operationalized. Similarly, in this paper, I have shown that the findings would only be limited to the African immigrants in California and those who are either smoking or have been diagnosed to have asthma.
Ethical Procedures
Any study that involves human participants is often subject to several ethical issues (Ndebele et al., 2014; Nicholls et al., 2015). The CHIS dataset used in this paper, as the main data for review, is subject to these ethical issues. In this section of the study, I will explain how these ethical issues affect the treatment of data and the recruitment of participants and materials when formulating the research findings.
The ethical procedures used by the CHIS researchers to collect data were approved by UCLA (2017). The ethical procedures that will be followed in the current study will also be based on the guidelines outlined by the Institutional Review Board (IRB) because I will apply to this institution to analyze data before conducting the statistical analysis. The current study will also involve an analysis of de-identified data to avoid cases of privacy infringement or confidentiality breaches. Data was stored safely in a computer and protected by a password, which only the researcher could gain access to. This safeguard ensured that no other person had access to the information besides the researcher (Nursing and Midwifery Board of Australia, 2012). Lastly, a credit will also be given to the original authors of the research to avoid cases where the readers could assume that the findings of the current study are the authors. This way, cases of plagiarism are avoided.
Summary
This chapter shows that the methodology of this research was intended to answer the main research question, which centered on investigating the association between asthma and smoking status among adult African immigrants in California. The independent variable is smoking and the dependent variable is asthma. The moderating factors are age, sex, years since immigration, marital status, alcohol use, education level, income level, and employment status. Secondary analysis of the archived data will be conducted after receiving IRB approval. I will conduct both descriptive and inferential statistical analyses and complete a descriptive analysis using frequency tables and measures of central tendency. I will also conduct the inferential analysis using bivariate (scatter plot diagram) and multivariate methods (MANOVA technique).
The quantitative correlation research is the main design used to answer this research question and it is based on the understanding that the research topic is a correlation study because the current research explores the association between asthma and smoking status among African immigrants in California. The main source of information is secondary research data, which comes from an independent survey by the CHIS.
The SPSS software tool is also the main data analysis technique used in this study because of its ability to analyze large amounts of data. The correlation technique is the main SPSS tool applied in this study because of its ability to identify associations between two or more variables. The results of the data analysis process are presented in the next chapter, which outlines the results and findings of the study
References
Arkkelin, D. (2014). Using SPSS to understand research and data analysis. Web.
Armijo-Olivo, S., Stiles, C., Hagen, N., Biondo, P., & Cummings, G. (2012). Assessment of study quality for systematic reviews: A comparison of the Cochrane collaboration risk of bias tool and the effective public health practice project quality assessment tool: Methodological research. Journal of Evaluation in Clinical Practice, 18(1), 12–18. Web.
Chan, A., Tetzlaff, J., Altman, D., Laupacis, A., Gøtzsche, P., & Krleža-Jerić, K. (2013). SPIRIT 2013 statement: Defining standard protocol items for clinical trials. Annals of Internal Medicine, 158(3), 200–207. Web.
Cheng, H. G., & Phillips, M. R. (2014). Secondary analysis of existing data: Opportunities and implementation. Shanghai Archives of Psychiatry, 26(6), 371–375. Web.
Faber, J., & Fonseca, L. M. (2014). How sample size influences research outcomes. Dental Press Journal of Orthodontics, 19(4), 27–29. Web.
Fernandez-Hermida, R., Calafat, A., Becoña, E., Tsertsvadze, A., & Foxcroft, D. (2012). Assessment of generalizability, applicability and predictability (GAP) for evaluating external validity in studies of universal family-based prevention of alcohol misuse in young people: Systematic methodological review of randomized controlled trials. Addiction, 107(9), 1570–1579. Web.
Hassan, S., Yellur, R., Subramani, P., Adiga, P., Gokhale, M., & Iyer M. (2015). Research design and statistical methods in Indian medical journals: A retrospective survey. PLoS ONE, 10(4), 12-18. Web.
Hoffmann, T., Bennett, S., & Del Mar, C. (2013). Evidence-based practice across the health professions. Chatswood, AU: Churchill Livingstone. Web.
Jain, S., Basavaraj, P., Singla, A., Singh, K., Kundu, H., Vashishtha, V., … Malhi, R. (2015). Bibliometric analysis of journal of clinical and diagnostic research (dentistry section; 2007-2014). Journal of Clinical and Diagnostic Research, 9(4), 47–51. Web.
Jirojwong, S., Johnson, M., & Welch, A. (2014). Research methods in nursing and midwifery. Sydney, AU: Oxford University Press. Web.
Lang, T. A., & Altman, D. G. (2014). Guidelines for reporting health research: A user’s manual. Oxford, UK: John Wiley & Sons, Ltd. Web.
Lee, C., Crawford, C., & Wallerstedt, D. (2012). The effectiveness of acupuncture research across components of the trauma spectrum response (tsr): A systematic review of reviews. Systematic Reviews, 1, 46-52. Web.
Liamputtong, P. (2013). Research methods in health. South Melbourne, AU: Oxford University press. Web.
Mathie, R., Roniger, H., & Van Wassenhoven, M. (2012). Method for appraising model validity of randomised controlled trials of homeopathic treatment: Multi-rater concordance study. BMC Medical Research Methodology, 12, 49-52. Web.
Mertz, M., Inthorn, J., Renz, G., Rothenberger, L., Salloch, S., Schildmann, J., … Schicktanz, S. (2014). Research across the disciplines: A road map for quality criteria in empirical ethics research. BMC Medical Ethics, 15, 17. Web.
Mohini, P., & Prajakt, J. (2012). What to use to express the variability of data: Standard deviation or standard error of mean? Perspective Clinical Research, 3(3), 113–116. Web.
Murari, G. (2013). Clinical research and its condition in India. Journals of Clinical Research in Bioethics, 4, 146-148. Web.
Ndebele, P., Wassenaar, D., Benatar, S., Fleischer, T., Kruger, M., Adebamowo, C., … Meslin, M. (2014). Research ethics capacity building in Sub-Saharan Africa: A review of nih fogarty-funded programs 2000–2012. Journal of Empirical Research on Human Research Ethics, 9(2), 24–40. Web.
Nicholls, S., Hayes, T., Brehaut, J., McDonald, M., Weijer, C., Saginur, R., & Fergusson, D. (2015). A scoping review of empirical research relating to quality and effectiveness of research ethics review. PLoS ONE, 10(7), 15-20. Web.
Nieswiadomy, R. M. (2012). Foundations of nursing research. Boston, MA: Pearson. Web.
Nursing and Midwifery Board of Australia. (2012). Code of professional conduct for nurses in Australia. Web.
Perry, Z., Barak, A., Neumann, L., & Levy, A. (2014). Computer-based learning: The use of SPSS statistical program for improving biostatistical competence of medical students. Journal of Biomedical Education, 1(3), 1-9. Web.
Polit, D. F., & Hungler, B. P. (2013). Essentials of nursing research: Methods, appraisal, and utilization. Philadelphia, PA: Wolters Kluwer. Web.
Rezaei, K., Klint, E., Hammer, H., Terslev, L., D’Agostino, M., Kisten, Y., … Arnaud, L. (2017). Analysis of correlation and causes for discrepancy between quantitative and semi-quantitative doppler scores in synovitis in rheumatoid arthritis. Rheumatology, 56(2), 255-262. Web.
Shaughnessy, J., Zechmeister, B., & Zechmeister, S. (2014). Research methods in psychology. Web.
Tamimi, A., Serdarevic, D., & Hanania, N. (2012). The effects of cigarette smoke on airway inflammation in asthma and COPD: Therapeutic implications. Respiratory Medicine, 106, 319–328. Web.
UCLA. (2017). California Health Information Survey (CHIS). Web.
Wright-St Clair, V., Reid, D., Shaw, S., & Ramsbotham, J. (2014). Evidence-based health practice. South Melbourne, AU: Oxford University Press. Web.
Appendix
Appendix 1: Data Analysis Matrix Table
Appendix 2: Mock Tables
Descriptive Statistics
- Objective: I.
- Procedure: Tabulation counts, percentages.
- Objective: Describe incidence of smoking.
- Procedure: Tabulations, Percentages.
- Objective: Describe incidence of African immigrants who have asthma.
- Procedure: Tabulations, Percentages.
- Objective: Describe number of African immigrant population in California.
- Procedure: Tabulations, Percentages.
Inferential Statistics
- Objective: Understand the Relationship between Asthma and Smoking.
- Procedure: Bivariate Analysis.
- Objective: Relationship between moderating, independent and dependent variables.
- Procedure: Multivariate analysis.