Nurses who are engaged in evidence-based practice are called upon to have a foundational knowledge of statistics to be able to analyze research into several health issues. In essence, it is critical for nurses to recognize the main statistical tests, when and how they are used and their implications in the research studies. An analysis into several chapters of Polit and Beck’s, book, Nursing research: Generating and assessing evidence for nursing practice reveals the basic statistical concepts a nurse should know.
Chapter 16: Descriptive Statistics
The chapter introduces descriptive statistics, which describes data characteristics. The chapter highlights different levels of scales such as nominal, ordinal, interval and ratio measurements. Additionally, the chapter defines frequency distributions, as well as distinguishing between univariate and bivariate descriptive statistics. Nominal and ordinal scales are qualitative and they label values such as ‘male’ ‘female’ ‘not satisfied’, ‘satisfied’ ‘very satisfied’ (Polit and Beck, 2017). Interval scales and ratio scales, which are numeric, measure both the order and the exact difference between values such as time and temperature, height and weight. Ratio scale, unlike interval scales, allows for an absolute zero. Frequency distribution measures the central tendency of data and includes mean mode and median. In essence, descriptive statistics determine how variables relate in terms of characteristics and distribution (Vetter, 2017). For example, a nurse is able to determine how many males or females were involved in a particular research study.
Delegate your assignment to our experts and they will do the rest.
Chapter 17: Inferential Statistics
The chapter focuses on inferential statistics, which describe populations based on the research sample. In particular, the chapter focuses on sampling distributions, levels of significance, and estimates of parameters. Moreover, the chapter outlines the most common inferential statistical tests such as ANOVA, t-tests and chi-square tests. On sampling distributions, researchers can use random sampling, purposive sampling among others to select sample group (Polit and Beck, 2017). Estimates of parameters measure how results deviate from the population mean. ANOVA measures the differences between two means, t-tests compares the means of two populations while chi-square tests assess the effectiveness of the fit between a set of observable and theoretical values (Vetter, 2017) .In essence, these tests are used in populations with diverse and unrelated characteristics.
Chapter 18: Multivariate Statistics
The chapter focuses on multivariate statistics, which are used in analyzing data with more or three variables. The chapter specifically describes simple linear regression, multiple linear regression, analysis of covariance (ANCOVA), causal modeling, and event history analysis. Simple linear regression establishes the relationship between the independent and dependent variable. Multiple linear regressions describe the connection between one continuous and dependent variable with two or more independent variables. Analysis of covariance (ANCOVA examines the differences and influence of the dependent variables means in relation to the controlled independent variables (Vetter, 2017). Causal modeling examines how one or more dependent variables exhibit a causal relationship with one or more of the independent variables. Event history analysis is statistical methods used in the description, explanation, and prediction of the occurrence of events (Windle, et al., 2018). The multivariate analysis was employed in the research study to establish adverse childhood health behaviors and experiences as well as outcomes among college students.
Chapter 19: Processes of Quantitative Data Analysis
This chapter focuses on analyzing and interpreting numerical results following a successful collection of research data to come up with meaningful insights. There are several steps in quantitative data analysis. The first step is data validation where a researcher finds out whether the data was collected as per the set standards and without any bias. Next, the researcher edits data to do away with common data collection errors. From here the researcher codes the data by grouping and assigning values to the survey responses to make data representation easier (Polit and Beck, 2017). All these steps ensure that the data is credible and is a true representation of the research process and outcomes.
Chapter 24: Qualitative Data Analysis
This chapter focuses on the techniques that researchers use to analyze qualitative data which Is more complicated that quantitative data. In essence, there are no specific tests to be used like in quantitative data as it is comprised of observations, words, symbols, and images. Instead, the analysis of qualitative data depends on the researcher’s capacity to recognize patterns of behavior and similarities in the events. However, a researcher should follow several steps in the analysis if qualitative data. The first step is data preparation where the research gets familiar with the data to come up with patterns (Vetter, 2017). Next, the researcher revisits the research objectives to establish which questions are to be answered. After this, the researcher develops a framework to identify broad concepts, ideas, and phrases for coding. From here, the researcher identifies connections and patterns to answer the research questions and find opportunities for further research.
References
Polit, D. F., & Beck, C. T. (2017). Nursing research: Generating and assessing evidence for nursing practice (10th ed.). Philadelphia, PA: Wolters Kluwer.
Vetter, T. (2017). Descriptive statistics: Reporting the answers to the 5 basic questions of who, what, why, when, where, and a sixth, so what? Anesthesia & Analgesia, 125 (5), 1797–1802.
Windle, M. et al. (2018). A multivariate analysis of adverse childhood experiences and health behaviors and outcomes among college students. Journal of American College Health, 66 (4), 246–251.
