explain why hierarchal multiple regression is used rather than parametric multiple regression analysis
Using the attached document in 200 words or more explain why hierarchal multiple regression is used rather than parametric multiple regression analysis.
Running Head: DOCTORAL STUDY PROPOSAL 1
DOCTORAL STUDY PROPOSAL 2
Doctoral Study Proposal
Professor’s Name
Student’s Name
Course Title
Date
Prospectus: The Impact of Transformational Leadership on Organizational Performance and Employee Retention
Problem Statement
Employee retention continues to remain a complex challenge in small and medium-sized enterprises (Park et al., 2019). In 2018, employee turnover was estimated to be 19%, where 77% of the rate was preventable by leaders through a change of their leadership styles (Anitha & Farida, 2016). According to Mcfeely and Wigert (2019), the cost associated with the replacement of an employee can be as low as half and as high as twice the annual salary of the employee. Thus, in an organization consisting of 100 employees with an average annual salary of $50,000, the cost of turnover and replacement could range from $660,000 to $2.6 million each year (Mcfeely & Wigert, 2019). In this regard, the leadership style adopted in an organization plays a significant role in shaping the effectiveness of that organization (Jiang, Zhao, & Ni, 2017). Leadership style influences not only employee retention but also the performance of the organization (Covella, McCarthy, Kaifi, & Cocoran, 2017), as employee retention has been found to affect both the performance and longevity of an organization (Nelms, 2018). Researchers have noted that low employee retention in organizations could be managed through components of transformational leadership, which could improve both the organizational performance and employee retention (Effiong et al., 2017; Tian et al., 2020). However, there is a lack of literature in which transformational leadership components have been examined in relation to both employee retention and organizational performance. The general business problem is the high rate of turnover in organizations and its subsequent impact on organizational performance (Boamah, Laschinger, Wong, & Clarke, 2018; Tian et al., 2020). The specific problem is the need to understand the relationship between transformational leadership components (idealized influence attributed, idealized influence behaviors, inspirational motivation, intellectual stimulation and individualized consideration) and employee retention and organizational performance.
Purpose Statement
The purpose of this quantitative correlational study is to examine the relationship between transformational leadership components (idealized influence attributed, idealized influence behaviors, inspirational motivation, intellectual stimulation, and individualized consideration) and organizational performance as well as employee retention. The target population will consist employees at a business organization located in Colorado Springs, Colorado. The independent variables of the study consist of five transformational leadership components, namely idealized influence attributed, idealized influence behaviors, inspirational motivation, intellectual stimulation, and individualized consideration. The dependent variables consist of organizational performance, to be measured through percentage of annual sales performance goal that is met by the employees; and employee retention, to be measured through the reported willingness of employees to stay in the organizations longer. It is expected that the findings of the study will help provide indicators for improvements in organizational performance and turnover rate through the adoption of transformational leadership components by senior leaders at small and medium-sized enterprises. Additionally, in helping improve employee retention and performance at small and medium-sized enterprises, the findings could help create employment in the region under study, reducing the negative social and individual consequences of unemployment.
Participants
The target population of this study will consist of employees who work at small to medium-sized enterprises in Colorado Springs, Colorado. The sample size will be determined based on a power analysis will be conducted through G*Power. Effective sizes across three categories, namely small, medium, and large for the purpose of hypotheses testing will be determined. Settings for G*Power will rely on hierarchal multiple regression test will be set to examine the relationship between transformational leadership components and organizational performance as well as employee retention.
Data will be collected using three sources. First, with regards transformational leadership components, a survey will be used in which all five components (idealized influence attributed, idealized influence behaviors, inspirational motivation, intellectual stimulation, and individualized consideration) are included. For this purpose, I will use Multifactor Leadership Questionnaire (MLQ) (Avolio and Bass, 1995), which consists of 20 questions across the five components with a five-point Likert-like scale for participants to provide their responses. Regarding the dependent variable of employee retention, I will use a five-item scale for employee retention measurement by Kyndt et al. (2009). Regarding the dependent variable of organizational performance, which will be measured through percentage of annual sales performance goal that is met by the employees, I will obtain financial performance documents from the organization for analysis.
Participants will be contacted through their organization’s HR executives. Participants selected in the study will be asked to fill a questionnaire survey consisting of the two instruments for measuring transformational leadership components and employee retention. It will be ensured beforehand that participants are educated regarding their rights to confidentiality and anonymity. Each questionnaire survey will take approximately 45 minutes, and participants will be given the opportunity to fill the survey either at home or at work. Organizational performance reports will be obtained from the HR executive for the latest financial year.
Regarding analysis, to ensure suitability of the data for statistical analysis, I will conduct tests of the following assumptions for multiple linear regression: standard residuals, multicollinearity, independence of errors, and variance and linearity. Measures will be examined through internal consistency coefficients, standard deviations, and means through the use of Cronbach’s, which will have an alpha of .01. Data will be imported to the statistical analysis tool SPSS. Next, I will conduct a Pearson’s product moment correlation in order to measure whether there is a relation between transformational leadership components and employee retention. Similarly, the relation between transformational leadership components and organizational performance will be assessed. I will use Pearson’s correlation r and hierarchical regression analysis to examine the relation between the variables.
Research Method
The research will adopt a quantitative methodology. Quantitative research would be used to measure and analyze the numerical data that would be collected using questionnaires. The data collected would be used to make a generalization on the impact of transformational leadership on organizational performance and employee turnover intention. The study will establish statistically the percentage of employees that are willing to leave or stay in their current organization based on the character traits, behavior, motivations, influence, and the inspiration they draw from their leaders. The descriptive study would provide data that would be highly reliable since the process can be repeated to verify it.
Quantitative research method would be appropriate for this study because it helps determine to what extent the independent variable affects the dependent variable. It is applicable in the aspects of social behavior that can be represented numerically (Nardi, 2018; Togao et al., 2018). Various instruments such as questionnaires are used in the collection of data, and statistical methods are applied in the analysis of the collected data. The collected data is analyzed and presented in charts, tables, and columns to help the researcher make conclusions. The researcher is able to interpret the meaning of collected data and establish associations and frequencies.
The main advantages of quantitative research methods are that the results can be easily generalized to represent the characteristics of the whole population since a larger sample is used. Besides, the process of data analysis takes less time as compared to qualitative research because statistical software like SPSS is used. According to Powers and Powers (2015), large samples make the research findings more truth-worthy. Also, reliability is maintained by the use of two or more independent variables (Carroll & Bailey, 2016; Brannen, 2017; Kumar, 2019). Carroll and Bailey (2016) conducted research to assess language proficiency where several independent variables were used, including listening, reading, writing, and speaking, citing that the variables significantly influence language proficiency.
The major disadvantage of quantitative research is that it only relies on data collected at a specific point ant time which may be influenced by an individual’s mood at the time of the interview (Brannen, 2017; Kumar, 2019). However, the lack of direct connection with the participant means that the researcher is unable to dig deep to obtain more information and only takes a snapshot of the phenomenon. Therefore, it only gives an overall picture of the population without looking into the underlying explanations and meanings.
Research Design
The study would employ a correlational research design. In this case, the researcher seeks to investigate the extent to which the character traits, behavior, motivations, influence, and inspiring leaders give their employees influence employee turnover intention and organization performance. As a correlational study, it would examine the type of relationship that exists between the variables (Brannen, 2017; Kumar, 2019). The study would show the relationship between leaders’ leadership styles and the employee turnover intention by use of techniques such as correlations, and cross-tabulation. The study would establish a regression equation that would be used to generalize the characteristics of the relationship between the variables and allow the researcher to make predictions on future trends.
Statistical analysis allows the researcher to measure the relationship between the variables and determine the direction and degree of the relationship. The degree of the relationship between shows how closely related the variables are, which defines whether there is a significant relationship between the variables (Curtis et al., 2016; Nardi, 2018). Correlation coefficients are used in correlational studies which range between -1 and +1. When there, the correlation is equal to zero, it means that there is no relationship between the variables. In this case, the null hypothesis is confirmed. If the correlation coefficient is either -1 or +1, there if perfect correlation. In this case, there is a significant relationship between the variables with -1 representing a negative relationship and +1 representing the positive relationship. Correlational research doesn’t imply causation like the experimental design. Therefore, there is a need for further research to determine whether the independent variables have a causal effect on the behavior of the dependent variable.
Population and sampling
Population
The target population of this study will consist of employees who work at small to medium-sized enterprises in Colorado Springs, Colorado. Participants for the study will be selected from frontline supervisors, sales professionals, mid level managers and chief executive officers working at the mid sized enterprises. The target population will have extensive experience in business. This will allow them to provide significant insights concerning how transformational leadership impacts organizational performance and employee retention.
Sampling
Non-probabilistic sampling will be used in this study. Particularly, purposive sampling will be used in this study to select the sample for the study. This approach is used since the researcher is seeking certain individuals in the identified population having specific characteristics. There are numerous strengths associated with purposive sampling. This type of sampling can be applied in a wide range of research designs. It will also allow the researcher to collect the most appropriate data to draw conclusion from. In addition, it allows for generalization of the results obtained. It provides the justification required for the researcher to make a generalization (Sharma, 2017).
However, there are various weaknesses associated with this sampling method. It introduces a lot of inferential statistics that can complicate the analysis process. It also introduces researcher bias that can affect the validity of the study. This is true regardless of the method used to collect data (Sharma, 2017). The participants used in the study can also manipulate the data being collected. People can initiate a change in their behavior after realizing that they have been selected for the study. They would provide responses that allow the researcher to reach at their preferred conclusions. Some participants may provide untrue responses to influence the outcome of the study. However, the strengths of purposive sampling outweigh its limitations (Sharma, 2017).
Non-probabilistic sampling is the most appropriate for this study because the study will use co-relational research design which is non-experimental. Non- probability sampling is used in non-experimental designs where the outcomes expected have already occurred (Meng et al., 2019. In addition, this method of sampling is aligned to this study as it will allow the researcher to generalize the results to the general population. This is important because the study aims to determine what happens to the general population regarding how transformational leadership impacts employee retention and organizational performance (Meng et al., 2019).
The sample size will be determined based on a power analysis. This will be achieved through G*Power. Effective sizes across three categories, namely small, medium, and large for the purpose of hypotheses testing will be determined. Settings for G*Power will rely on hierarchal multiple regression test will be set to examine how transformational leadership components impacts organizational performance as well as employee retention among small and medium organizations.
Employees who have worked for their respective organizations for more than three years will be included in the study. Participants of the study must also be working at small to medium-sized enterprises in Colorado Springs. Individuals who are experiencing mental health issues will be excluded in the study. In addition, employees who have not attained the age of 30 years will also be excluded from the study. Employees who are still in the university will also be excluded from the study.
Ethical Research
Before the research begins, participants will be required to fill a consent form. Sufficient information regarding the study will be provided to ensure that they understand the implications of taking part in the study. Information regarding the duration of the study, nature and purpose of the study, benefits and probable risks will be included in the consent form. While the consent form will be written in simple English, the researcher will try to explain certain areas for the participants to have a better understanding of the research (Ozdil, Seneviratne & Mai, 2017). Participants will also be educated regarding their rights to confidentiality and anonymity. After reading the form and understanding the information provided, participants will be required to voluntarily provide their willingness to take part in the study by signing the form (Ozdil, Seneviratne & Mai, 2017).
Participants will be allowed to end their involvement in the study at any stage if they feel that they cannot continue participating. At the time of withdrawal, study subjects will be required to inform the research team that they intent to withdraw. The participants will be provided with a withdrawal template with certain sections to fill. In the form, the participants may provide reasons for withdrawal but that is not compulsory. Such participants may also provide recommendation on how the study can be improved. There will be no penalties or loss of benefits to which the participants are entitled if they discontinue participation.
Participants will be rewarded with gift cards for taking part in the study. The incentives will be provided at the end of the study participation. This implies that after a participant fills the form of withdrawal then he/she is entitled to receive the incentives. The reward will accrue as the study progresses and will not be contingent upon the respondents completing the study. Gift cards are an effective way of encouraging individuals in the target population to take part in the study. It also presents an easy way for the researcher to reward the participants for their time and effort put in the study (Ozdil, Seneviratne & Mai, 2017).
The researcher will discuss with participants the level of confidentiality and privacy associated with the study. Participants will not be exposed to risks higher than those experienced in their normal lives. Participants will be protected from mental and physical harm avoiding offending, frightening or embarrassing them (Ozdil, Seneviratne & Mai, 2017). The information obtained in the study will be stored safely in the researcher’s office for five years. During this period the information will only be used for study purposes. After five year the questionnaires will be burned to prevent the information from landing in the wrong hands. Participants will not be required to states their names or their organizations while filling the questionnaire survey (Ozdil, Seneviratne & Mai, 2017). This will ensure confidentiality and privacy of their personal information.
Data Collection-Instruments
Three data collection instruments will be used for this study. First, with regards transformational leadership components, a survey will be used in which all five components (“idealized influence attributed, idealized influence behaviors, inspirational motivation, intellectual stimulation, and individualized consideration”) are included. For this purpose, Multifactor Leadership Questionnaire (MLQ) which consists of 20 questions across the five components with a five-point Likert-like scale for participants will be used to provide their responses (Van Jaarsveld, Mentz & Ellis, 2019). Regarding the dependent variable of employee retention, a five-item scale will be used for employee retention measurement by.
Regarding the dependent variable of organizational performance, which will be measured through percentage of annual sales performance goal that is met by the employees, financial performance documents will be obtained from from the organization for analysis. Ordinal scale will be used for measuring the five transformational leadership components listed above. The MLQ is the most appropriate instrument for this study as they are inexpensive and convenient as they allows the study participants to provide answers either at home or in their work places.
Each questionnaire survey will take approximately 45 minutes, and participants will be given the opportunity to fill the survey either at home or at work. Participants will choose between filling online questionnaires and filling them on paper. Those who will choose to fill on paper will require pencil to fill the questionnaire. Participants who choose to fill the survey online will require any digital device such as a mobile phone, laptop or desktop computer. The questionnaires will be sent to them through emails. Participants will be required to tick on the most appropriate box for the various transformational leadership components provided.
Construct validity will be used to measure the validity of the MLQ. Confirmatory factor analysis will be the specific method used to assess how the questionnaire items are aligned to the theoretical model. Content validity will be used in the study to assess whether all aspects of the construct are represented (ERGİN & AKIN, 2018). This will involve testing the relevance of the survey questions. Criterion based validity will be used to measure how transformational leadership impacts employee retention and organizational performance. Convergent/divergent validation will be used to determine how the study variables are related (ERGİN & AKIN, 2018).
Internal consistency will be used to measure how the items in the likert scale measure different aspects of the same transformational leadership components. If all components in the test measure the same idea then the test will have internal consistency reliability. Expert panels will be used to validate the survey questions for the purposes of enhancing the reliability and validity of the study. Triangulation will be used in data analysis to enhance the reliability of the study. Raw data for this study will be provided in tables. The tables will be attached in appendix B of this paper.
Data Collection Technique
Questionnaire surveys will be used for data collection in this study. Participants will be contacted through their organization’s human resource executives. Participants selected in the study will be asked to fill a questionnaire survey consisting of the two instruments for measuring transformational leadership components and employee retention. It will be ensured beforehand that participants are educated regarding their rights to confidentiality and anonymity. Each questionnaire survey will take approximately 45 minutes, and participants will be given the opportunity to fill the survey either at home or at work. Organizational performance reports will be obtained from the HR executive for the latest financial year (Taherdoost, 2016).
Survey is the most appropriate data collection technique for this study due to the numerous benefits associated with it. It is more practical and cost effective as compared to other quantitative techniques of data collection. It is also easy to build and implement as compared to other complex statistical techniques used in quantitative studies. In addition, surveys allow the researcher to collect information within a very short period of time. In this particular study, participants will be expected to complete the questionnaire within 45 minutes. It is also more convenient for the respondents as they can fill the questionnaire either at home or in their workplace.
The survey will also be administered in small and medium businesses in Colorado Springs preventing geographical dependence. The researchers will be able to access the participants easily. It will also allow the researchers to collect data from a large number of participants. This is important as it will allow for the generalization of the results. Additionally, the survey will allow for the use of triangulation to enhance the reliability and validity of the research. It will also reduce the errors associated with other data collection approaches. The survey will also be administered through emails depending on the choices of the participants (Taherdoost, 2016).
The technique will also allow broad data to be collected. This is important in determining how transformational leaderships impacts employee retention and organizational performance. It will also allow the researcher to ask a wide range of questions regarding both the dependent and independent variables giving adequate flexibility in the process of data analysis. It will also be way to administer the questionnaire. Finally, due to the high representativeness associated with the technique, the researcher will be able to find statistically significant results as compared to other techniques.
However, there are various limitations associated with survey. The technique has an inflexible design. The survey that was use from the start can’t be altered in the process of data collection. However this weakness can be seen as strength since fairness and preciseness can both be applied in the research. The technique is also associated with possible inappropriateness of questions. Dues to standardization, the researcher may have to develop general questions to take care of the entire population. However, these questions may be inappropriate for some respondents (Taherdoost, 2016).
In addition, respondents may not feel motivated enough to provide accurate answers. This limitation has been addressed by providing incentives to the participants to encourage them provide truthful answers. The survey may also have some errors. To address this limitation, the survey questionnaire will be reviewed by experts before being administered. A lot of assumptions may be made in the selection of participants affecting the credibility of the study. A pilot study will be conducted using the process described above after IRB approval.
Statistical analysis will be used for analyzing the data collected. Pearson’s correlation and hierarchical regression analysis will be used to examine the relation between the variables in this study. This approach will allow the researcher have a better predictive capacity as compared to other methods (Ong & Puteh, 2017). The approach allows the researcher to measure the relationship between the variables and determine the direction and degree of the relationship. The degree of the relationship between shows how closely related the variables are which defines if there is a considerable relationship between the variables. Correlation coefficients are used in co-relational studies which range between -1 and +1.
When there, the correlation is equal to zero, it means that there is no relationship between the variables. In this case, the null hypothesis is confirmed. If the correlation coefficient is -1 or +1 then there is perfect correlation. In this case, there is a significant relationship between the variables with -1 representing a negative relationship and +1 representing the positive relationship (Ong & Puteh, 2017). Co-relational research doesn’t imply causation like the experimental design. Therefore, there is a need for further research to determine whether the independent variables have a causal effect on the behavior of the dependent variable (Ong & Puteh, 2017).
Other statistical methods such as two-way Anova are not appropriate for this study because it does not align with the nature of the study questions. The method is too complex making it difficult for the researcher to draw effective conclusions for the study (Ong & Puteh, 2017). In addition, the approach also tests for means and the researcher cannot be able to use it to establish which mean is different from the other. In this particular study, the researcher may be unable to evaluate the relationship between the dependent and independent variable (Ong & Puteh, 2017). This could make it difficult to draw viable conclusions for the study. hierarchical regression analysis is therefore the most appropriate approach that allows the researcher to determine how transformational leadership impacts employee retention an organizational performance.
Regarding analysis, to ensure suitability of the data for statistical analysis, tests of the following assumptions for multiple linear regression will be conducted: standard residuals, multi-collinearity, independence of errors, normality and variance and linearity. Measures will be examined through internal consistency coefficients, standard deviations, and means through the use of Cronbach’s, which will have an alpha of .01. Data will be imported to the statistical analysis tool SPSS. Next, a Pearson’s product moment correlation will be conducted in order to measure whether there is a relation between transformational leadership components and employee retention. Similarly, the relation between transformational leadership components and organizational performance will be assessed.
The assumptions outlined above may be violated. This may have significant impacts on the validity of the results. However, various actions will be taken to minimize the consequences of violating the assumptions. One of these actions is data transformation. This action will be taken particularly if the assumption of normality is violated. Data transformations such as square root or natural log transformations will be used. However, only the transformed variable can be interpreted after data transformation. As a result, the researcher will not be able to provide interpretations based on untransformed values.
If data transformation will not be able to solve the issue or the researcher encounters multiple violations of the assumptions, non-parametric analysis will be done. Ordinal data will be used to take this action. This actions done not assume that the data is obtained from a normal distribution (Sheard, 2018). Rather, statistical measurements are used to estimate the distribution. As a result, this action will only be taken if the data obtained is not normal. This action has however been found to be less powerful as compared to parametric analysis. Alternative statistics will also be used to determine the significant is certain assumptions are violated (Ong & Puteh, 2017).
The data obtained will be screened to avoid any mistakes that could have been made in data entering process. This will involve checking out abnormal data from the frequency tables. The researcher will also go back to the questionnaires and make appropriate corrections. The data will also be cleaned to identify missing data and normality. Missing data will be detected by from the frequencies of the study variables. The data will then be sorted in descending and acceding orders. The missing values are identified at this point and the researcher will go back and fill them again. If the missing values are less than ten percent then the data will be substituted with imputed value or neural value (Sheard, 2018).
Inferential results will be interpreted by estimating key parameters. Random sample will be obtained from the data and estimated. Point estimates will be used to establish the relationship between the variables of the study. Confidence intervals will also be used to interpret inferential results (Sheard, 2018). This will allow the research team to obtain results and the data measurement level. It will also provide the strength as well as statistical significance and direction of the effect. Probability values will also be used to interpret inferential results. This will allow the researcher the opportunity to have a sample size that represents the general population. It will also allow the researcher to use confidence intervals effectively to validate the results (Sheard, 2018).
SPSS will be used for data analysis in this study. Numerous studies have indicated that SPSS is an effective tool for manipulating data obtained through surveys. Data obtained in surveys can easily be imported to SPSS. It is also easily customizable due to its flexibility. This can allow the researcher to get super granular data sets from the analysis. Minimal legwork will be required as the tools will automatically import variable types, names and titles (Ong & Puteh, 2017). Once data has been imported to SPSS, a Pearson’s product moment correlation will be conducted in order to measure whether there is a relation between the variables of the present study.
Study Validity
There are various threats to statistical conclusion validity that lead the researcher to make incorrect conclusions about the relationship between variables. The researcher can make two types of errors. First he/she can conclude that there is no relationship when in the actual sense it is there (Mirhaghi & Christ, 2016). Secondly the researcher can conclude that there is a relationship when it is not there. One of these threats is low statistical power. Having a low statistical power can lead the researcher to come into conclusion that there is no relationship between transformational leadership and employee retention and organizational productivity (Mirhaghi & Christ, 2016).
Violated assumptions for test are another significant threat to statistical conclusion validity. The inferential statistics in this study involve various assumptions aimed at making the analysis suitable for testing the hypothesis. Any violation of these assumptions can lead to incorrect conclusions about the relationship between variables. Another threat is Dredging. Each hypothesis test involves the alpha rate. The researcher may inflate the type I error by dredging or searching through the data testing various hypotheses to find a relationship. The more the data is tested the increased risk of observing a type I error and consequently making incorrect conclusions about the study variables.
Unreliability of measure is another significant threat to statistical conclusion validity. The researcher can make incorrect conclusions if the independent or the dependent variables are not measure reliably. Restriction of range is another considerable threat to conclusion validity. Restriction of range such as selection effects, ceiling effects and floor effects may introduce type I errors and reduce the power of the study. This is because reduced variability can attenuate correlations. Another threat is heterogeneity of the study units. High heterogeneity of participants may affect the study by obscuring true relationships between variables. There is also threat to internal validity that may lead to biased results and consequently result in incorrect inferences (Mirhaghi & Christ, 2016).
However, there are various approaches that will be adopted by the researcher to minimize the impacts of these threats. One way is to ensure that the researcher has extensive statistical knowledge. This will be achieved by formal statistical training to the researcher before the study. This will provide the researcher with significant insights on how to apply research methods reducing the threats. Another way is to increase the statistical power. This will be achieved by collecting more information. One way to do that is to increase the sample size for the study. This will help the researcher to avoid making incorrect inferences about the relationships between variables.
Another approach that will be used is to increase the probability that the researcher will make type I error. In other word, the researcher will increase the chances that there will be a relationship between variables when I the actual sense it is not there. This will be achieved by raising the alpha level. For example, instead of using 0.10 as the significance level, the researcher can use 0.15. This will ensure that the researcher will make correct inferences concerning the relationships between variables. Using a lower significance level will increase the chances that an incorrect inference will be made.
The researcher will also mitigate the threats to statistical conclusion validity by increasing the effect size. The effect size is a ration of the noise in the context to the relationships’ signal. As a result, the researcher will choose either to decrease the noise or up the signal (Mirhaghi & Christ, 2016). Decreasing the noise in this case implies improving the reliability of the study. On the other hand, the researcher will increase the salience of the relationship between the transformational leadership and dependent variables to up the signal. All these approaches will be used to improve the conclusion validity of this study (Mirhaghi & Christ, 2016).
Non-probabilistic sampling was used to select the sample size for this study. This sampling method hinders external validity for the study. As a result the findings for this study can only be generalized to small and medium sized businesses in Colorado Springs. This is because it is difficult for the researcher to determine how effective the general population has been represented in the study (Mirhaghi & Christ, 2016). However, the researcher will improve the external validity for this study by ensuring the withdrawal levels are low (Mirhaghi & Christ, 2016). A lot of information will also be provided concerning similarities between the locations and people in the study.
Transition and summary
In summary, the target population of this study will consist of employees who work at small to medium-sized enterprises in Colorado Springs, Colorado. Non-probabilistic sampling will be used in this study. Particularly, purposive sampling will be used in this study to select the sample for the study. This approach is used since the researcher is seeking certain individuals in the identified population having specific characteristics. Non-probabilistic sampling is the most appropriate for this study because the study will use co-relational research design which is non-experimental. Non- probability sampling is used in non-experimental designs where the outcomes expected have already occurred.
Before the study starts, those who wish to participate will be required to fill a consent form. Sufficient information regarding the study will be provided to ensure that the respodents understand the implications of taking part in the study. Participants will be allowed to end their involvement in the study at any stage if they feel that they cannot continue participating. At the time of withdrawal, respondents will be required to inform the research team that they intent to withdraw. Participants will be rewarded with gift cards for taking part in the study. The incentives will be provided at the end of the study participation.
Three data collection instruments will be used for this study including a survey, MLQ and a five-item scale. Questionnaire surveys will be used for data collection in this study. Participants will be contacted through their organization’s human resource executives. Statistical analysis will be used for analyzing the data collected. Pearson’s correlation and hierarchical regression analysis will be used to examine the relation between the variables in this study. The researcher will use various strategies to mitigate various threats to statistical conclusion validity. This study used non-probabilistic sampling which hinders eternal validity. The researcher will use various approaches to enhance external validity.
Section there will cover the literature review on the present topic. The literature to be reviewed in the section will be relevant to the research questions and problem statement. Historical background of the study will also be provided in this section. In addition, theoretical model of the study will be discussed extensively. Literature concerning, transformational leadership, organizational performance and employee retention will be reviewed in a logical manner. Concepts will be contrasted and compared to make on argument. Finally, a succinct summary documenting the main points of the section will be provided.
References
ERGİN, E., & AKIN, B. (2018). The Turkish adaptation of a quality assessment tool for quantitative studies: validity and reliability analyses. Türkiye Klinikleri Hemşirelik Bilimleri Dergisi, 10(4), 292-308. Retrieved from
https://cbgoguscerrahisi.turkiyeklinikleri.com/article/tr-the-turkish-adaptation-of-a-quality-assessment-tool-for-quantitative-studies-validity-and-reliability-analyses-82848.html
Meng, Z., Zhang, D., Li, G., & Yu, B. (2019). An importance learning method for non-probabilistic reliability analysis and optimization. Structural and Multidisciplinary Optimization, 59(4), 1255-1271. Retrieved from
https://link.springer.com/article/10.1007/s00158-018-2128-7
Mirhaghi, A., & Christ, M. (2016). The conclusion validity of the triage scales has been threatened. Notfall+ Rettungsmedizin, 19(4), 299-300. Retrieved from
https://link.springer.com/article/10.1007/s10049-016-0165-6
Ong, M. H. A., & Puteh, F. (2017). Quantitative data analysis: Choosing between SPSS, PLS, and AMOS in social science research. International Interdisciplinary Journal of Scientific Research, 3(1), 14-25. Retrieved from
https://www.researchgate.net/profile/Fadilah_Puteh/publication/322885790_Quantitative_Data_Analysis_Choosing_Between_SPSS_PLS_and_AMOS_in_Social_Science_Research/links/5a73d1320f7e9b20d490034b/Quantitative-Data-Analysis-Choosing-Between-SPSS-PLS-and-AMOS-in-Social-Science-Research
Ozdil, E., Seneviratne, C., & Mai, X. T. (2017). Ethical considerations in qualitative research. The Routledge companion to qualitative accounting research, 483. Retrieved from
http://alitsaki.ir/wp-content/uploads/2019/10/The_Routledge_Companion_to_Qualitative #page=508
Sharma, G. (2017). Pros and cons of different sampling techniques. International journal of applied research, 3(7), 749-752. Retrieved from
https://d1wqtxts1xzle7.cloudfront.net/58765080/Pros_and_cons_of_sampling ?1554128717=&response-content-
Sheard, J. (2018). Quantitative data analysis. In Research Methods: Information, Systems, and Contexts, Second Edition (pp. 429-452). Elsevier. Retrieved from
https://research.monash.edu/en/publications/quantitative-data-analysis
Taherdoost, H. (2016). Validity and reliability of the research instrument; how to test the validation of a questionnaire/survey in a research. How to Test the Validation of a Questionnaire/Survey in a Research (August 10, 2016). Retrieved from
https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3205040
Taherdoost, H. (2016). How to Design and Create an Effective Survey/Questionnaire; A Step by Step Guide. International Journal of Academic Research in Management (IJARM), 5(4), 37-41. Retrieved from
https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3224226
Van Jaarsveld, L., Mentz, P. K., & Ellis, S. (2019). Implementing the Multifactor Leadership Questionnaire (MLQ) in a challenging context. International Journal of Educational Management. Retrieved from
https://www.emerald.com/insight/content/doi/10.1108/IJEM-02-2018-0041/full/html