Quantitative Data Collection and Analysis

Subject:	Sciences
Pages:	8
Words:	2294
Reading time:	9 min
Study level:	PhD

Case Assignment

Introduction of a Quantitative Article

In the majority of cases, men are defined as more appropriate and beneficial examples of leaders than women. However, to disprove this theme and discuss this problem from a new perspective, the researchers introduce different methods and use different sources. For example, the article that was written by Paustian-Underdahl, Walker, and Woehr for the Journal of Applied Psychology in 2014 aimed at discussing the differences between male and female leadership styles using the meta-analysis research design. The results of this quantitative statistical analysis helped to extend the role of a theory and its appropriate application to male and female leadership (Paustian-Underdahl, Walker, & Woehr, 2014).

Research Design

In the article under discussion, a meta-analysis was chosen as the main research design with the help of which the integration of the results from different independent studies was developed (Salkind, 2010). The goal of the research design was to quantitatively summarize gender differences that could occur in the perceptions of leadership and its effectiveness in regards to the chosen sample and studies. With the help of the chosen research design, six hypotheses were developed and tested (Paustian-Underdahl et al., 2014).

Summary and Evaluation of Data Collection

Sampling strategy. The strategy developed by the authors of the article aimed at gathering primary studies between 1962 and 2011. Several keywords were used to find the studies, including “leadership performance” and “leadership effectiveness” (Paustian-Underdahl et al., 2014). Random sampling was used while reviewing the past data where each unit of the population could have the same probability of being selected in regards to the criteria offered.

Sample size. The perception of leadership effectiveness and gender roles were investigated across 99 independent samples from 95 different studies (Paustian-Underdahl et al., 2014). 58 journal publications, 30 unpublished dissertations, five books and six other additional sources were used in this meta-analysis. The sample size was from 10 to 60 leaders from the USA and Canada, and the mean size was 1.011 leaders (Paustian-Underdahl et al., 2014).

Instruments. Meta-analysis was used to answer the main research question if gender issues could influence the perceptions of leadership effectiveness. The inclusion criteria were developed on the basis of several instruments such as the manual search through different journals, the consideration of the results of conference proceedings occurred between 2010 and 2012, and emails sent to different listservs.

Ethics. In this particular article, the authors did not pay much attention to the discussion of ethical issues of research. The goal was to gather the material from the already offered studies and develop them in terms of a new statistical meta-analysis. Therefore, it is possible to consider the idea that the ethics was properly evaluated in terms of past experiences and projects used in this analysis. The study developed a call that was based on the impact of contextual moderators who developed the studies on the basis of the congruity theory and other additional theoretical frameworks which could be defined as appropriate to answer the offered research hypotheses.

Summary and Evaluation of Data Analysis

Hypothesis testing procedures. In the article, there were six main hypotheses, and the last hypothesis was divided into two sub-statements. The authors suggested using multiple methods for the analysis of the data. For example, the chi-square-based Q statistic and the 75% rule were two methods chosen by the moderators to test the data gathered. The regression analysis developed for a meta-analysis was also used to cover the details of the work.

Data preparation. To prepared an appropriate data analysis, several groups of moderators were introduced. There were several categorical moderator variables with the help of which different ideas were generated and introduced in regards to the perceptions chosen. Continuous moderators focused on the identification of male and female rates and the time of the study offered.

Statistical analysis. In addition to the chi-square-based Q statistic method and the 75% rule method, the categorical models were developed to create an appropriate equivalent to the evaluation and the explanation of variance discovered in the data.

Results. In the results section, the authors clearly explained what kind of work was done, and what achievements were made to discussed the already developed hypotheses. Due to a small sample size offered in the study, the Q test of homogeneity was used to estimate the population values and compare the results between male and female leadership. The moderator analysis consisted of several parts to answer on each hypothesis in a clear and comprehensive way.

For example, to test the first hypothesis, the weighted least squares analysis in SPSS/PASW 18 was used. It was not supported because male leaders were proved as more effective than women (Paustian-Underdahl et al., 2014). The second hypothesis was partially supported because male leadership was effective only in male-dominated organizations. Third and fifth hypotheses were also partially supported by a number of clear arguments being presented. In comparison to the fourth hypothesis that was not supported, the sixth hypothesis was the only one that initially supported by all moderators as the connection between gender differences and self-ratings was proved and defined as effective.

Validity and Reliability of Findings

To access the issues of research validity, it is necessary to clarify if all research procedures were appropriate, and if three aspects, such as generalization, measurements, and control of variables, were considered (Hammersley, 2008). Clear measurements were introduced in the study regarding the years of publication, the presence/absence of keywords, and the appropriateness of the studies to the chosen research topic. Each hypothesis was in need of an additional investigation and the clarification of the details. Validity was proved by the offered systematic approach in data collection and the description of the studies, their appropriate referencing, and examples.

Reliability defined the correlation between the test scores and test forms. Quantitative research was reliable due to the description of the procedures and the identification of the results obtained. Though not all studies chosen for research provided the required alpha coefficient for reliability, the chosen effect sizes helped to fix the situation and follow the offered procedure (Paustian-Underdahl et al., 2014).

Contributions to Knowledge

Topic. The chosen article added a lot to the knowledge about the topic of gender issues in such field as leadership and education. Though male dominance in leadership was evident, it was necessary to discuss what factors could contribute to such dominance and explain why this topic was of frequent debates and controversies.

Research problem. The main research problem raised in this study was connected with the necessity to explain what determined gender differences and if it was possible to change the results. Relying on the results, it was possible to prove that male dominance could not be neglected.

Research process. The working process demonstrated in the article helped to test several hypotheses at the same time and contributed knowledge considerably by means of using several tests for the analysis and evaluation.

SLP Assignment

Quantitative data analysis	Authors	Analysis intent	Variables	Statistic symbol	Procedures
T-test
Single sample	Mehdinezhad and Sardarzahi (2015).	To investigate teachers’ perceptions of leadership behavior and the decisions made by their principals.	There were five main variables in the analysis, which were model the way, enable other people to act, inspire a shared vision, challenge the process, and encourage the heart (Mehdinezhad & Sardarzahi, 2015).	Mean (X) and standard deviation (s).	The calculated mean was defined as more significant than the assumed mean. The teachers had to rate their principals in regards to the variables offered in one analysis, and principals rated themselves in terms of the same variables in another analysis.
Independent sample	Mehdinezhad and Sardarzahi (2015).	To compare the views and evaluations of teachers and principals regarding the chosen variables	There were four main variables in the analysis (one variable was excluded from this type of work), which were model the way, inspire a shared vision, challenge the process, and encourage the heart (Mehdinezhad & Sardarzahi, 2015).	Mean (X), standard deviation (s), and correlation coefficient (r)	The results of this t-test included the description of the existing difference between the views and opinions of teachers and their principals. In fact, it was proved that teachers’ ratings were lower than principals’ ratings
Paired sample	Rosch, Collier, and Zehr (2014).	To identify the differences between self assessments and teammate assessments that were developed in regards to leadership behaviors. The chosen data analysis method helped to calculate the effect sizes of means differences (Rosch, Collier, & Zehr, 2014).	Transformational leadership, transactional leadership, self-efficacy to lead, and motivation to lead (Rosch et al., 2014).	N, µ, and SD.	First, the test was conducted on pre-test to indicate measurable scores in transformational and transactional behaviors during self assessments. Then, the test was conducted on the basis of student post-scores and teammate score for two types of leadership.
ANOVA	Karadag and Bektas (2014).	To improve the analysis of data and check the existing differentiation according to one of the variables chosen for research (seniority).	In whole research, there were three main variables: gender, school type, and seniority. ANOVA was one of the methods used by the researchers to analyze one variable only (seniority) (Karadag & Bektas, 2014).	X, µ, f	It was necessary to evaluate the perceived leadership behaviors that were demonstrated by women administrators in regards to the question of seniority. First, the homogeneity value of variances was identified. As soon as no significant differences were observed in the chosen dimensions, it was concluded that homogeneous disturbance was observed in the variances.
Pearson Correlation	Uzonwanne (2015).	This method of data analysis was used to measure the existing relationship between the offered variables, the chosen leadership style, and the decision-making process.	There were two demographic variables in research, age, and gender. These two variables were found to correlate below 0.01 that proved the strong relations between them (Uzonwanne, 2015). In the product-moment correlation, the types of leadership styles and decision-making models were chosen as the main variables.	p	In this analysis method, the initial dimensions (initiating structure and consideration) were used to test the first hypothesis. The information was divided into the categories and demonstrated a positive correlation between the dimensions
Regression
Multiple	Malik, Aziz, and Hassan (2014).	A multiple regression technique was used to analyze the data and determine the impact of such factors as age, gender, qualification, and experience. It helped to predict the acceptance of leader.	There were several regression techniques used to investigate the variables gained the forms of predictors which were age, gender, qualification, rank, experience, and service and the variables used for the analysis of task structure, role ambiguity, stress, needs, and abilities (Malik, Aziz, & Hassan, 2014).	β, p, r	Two multiple regression analyses were developed in the study. The first analysis was developed to investigate the relation between the subordinates and the level of leaders’ acceptance. The second analysis focused on the situational factors and subordinates. Bother analyses helped to reject the null hypothesis of the study.
Hierarchical	Holloway (2012).	Hierarchical regression analysis was conducted to investigate the chosen organizational climate dimensions and identify the possible contribution of the theory to the variables mentioned in the hypotheses (Holloway, 2012).	There were four variables for consideration in this research: education level, age, gender, job rank, and tenure (Holloway, 2012).	β	All hierarchical procedures were used to examine the relations between variables and climate dimensions (Holloway, 2012). Clear explanations and the examples of behaviors proved the validity and reliability of findings. Organizational needs and respect to people turned out to be a crucial part of the paper.
Logistic	Cook and Glass (2014).	The conditional logistic regression analyses were used to test the offered hypotheses and consider each variable chosen.	In the study, there were such variables as the transition, the size of a company, the number of women in the field, the position of the appointee (external or internal to the company) (Cook & Glass, 2014). All these variables could be divided into two main groups: control variables and predictor variables. Both of these types were crucial for conditional logistic regression.	β, p	The predictor variables were regressed after the control over all variables was promoted. The analysis proposed a negative trend of return to such issue as equity. No statistical significance was discovered. The composition of the board of directors turned out to be a significant aspect. To prove the validity and reliability of research, the offered variables were properly measured and explained. For example, the size of a company was measured by the available total assets. It was also important to identify the number of women in the industry and create a basis for the analysis.

Discussion Question Assignment

“Three Studies on the Leadership Behaviors of Academic Deans in Higher Education” is a three-article dissertation that was written by Rebecca Brower in the Florida State University in 2013. This three-article dissertation was based on mixed methods to investigate leadership behaviors in higher education. Three articles used in the dissertation contained different methods of data collection and analysis. One of them was characterized by the quantitative nature and the offered social network methods. SPSS was used for the quantitative data analysis. In the quantitative article, the issues of gender, political skills, and social capital were discussed to support the hypothesis of gender similarities.

In this dissertation, the TAD format was followed: the introduction aimed at discussing the articles and their connection to the topic offered, each section was devoted to one particular article, and the concluding part focused on the discussion of the results and the worth of each article in terms of the topic chosen. The only question about the research design that could be posed is if the quantitative method could be regarded as one of the best and the most appropriate options to rely on while discussing the hypotheses.

References

Brower, R. (2013). Three studies on the leadership behaviors of academic deans in higher education. Web.

Cook, A., & Glass, C. (2014). Women and top leadership positions: Towards an institutional analysis. Gender, Work and Organization, 21(1), 91-103.

Hammersley, M. (2008). Assessing validity in social research. In P. Alasuutari, L. Bickman, & J. Brannen (Eds.), The SAGE handbook of social research methods. (pp. 42-54). Thousand Oaks, CA: SAGE.

Holloway, J.B. (2012). Leadership behavior and organizational climate: An empirical study in a non-profit organization. Emerging Leadership Journeys, 5(1), 9-35.

Karadag, E., & Bektas, F. (2014). Women administrators in education: Leadership behavior assessment according to teachers perceptions. Journal of Studies in Social Sciences, 7(1), 32-40.

Malik, S.H., Aziz, S., & Hassan, H. (2014). Leadership behavior and acceptance of leaders by subordinates: Application of path goal theory in telecom sector. International Journal of Trade, Economics and Finance, 5(2), 170-175.

Mehdinezhad, V., & Sardarzahi, Z. (2015). A study of the leadership behaviors reported by principals and observed by teachers and its relation with principals management experience. Efficiency and Responsibility in Education and Science, 8(2), 48-53.

Paustian-Underdahl, S.C., Walker, & Woehr, D.J. (2014). Gender and perceptions of leadership effectiveness: A meta-analysis of contextual moderators. Journal of Applied Psychology, 99(6), 1129-1145.

Rosch, D.M., Collier, D.A., & Zehr, S.M. (2014). Self-vs.-teammate assessment of leadership competence: The effects of gender, leadership self-efficacy, and motivation to lead. Journal of Leadership Education, 5, 96-124.

Salkind, N.J. (Ed.) (2010). Encyclopedia of research design. Thousand Oaks, CA: SAGE.

Uzonwanne, F. (2015). Leadership styles and decision-making models among corporate leaders in non-profit organizations in North America. Journal of Public Affairs, 15(3), 287-299.