# Understanding the Basics of Statistics

Statistics deals with the development of techniques for gathering, analyzing, presenting, and interpreting factual or empirical data. It is an interdisciplinary field that applies to virtually all areas of science and the various research questions in those fields lead to the development of new methods and theories. In studying the theories of statistics, researchers applying this method draw upon various computational and mathematical tools. Collection of statistical data happens in everyday life. People encounter many situations in science or in life in which outcomes are uncertain.

In some cases, there is uncertainty because a person has not yet determined the outcome, for example, one might not know whether there will be a storm the next day, while in other situations, there is uncertainty because even though the person has determined the outcome, people are not aware of it. For example, one might not know whether they passed an exam. To get certainty, therefore, the researcher analyzes the data using various analytical tools to achieve more informed information. Statistics is an essential part of research because researchers need to analyze the data they collect for them to present and interpret to others. Before we begin data collection, it is essential to have a clear plan for analysis that will offer a guide from the stage of summarization to testing hypotheses. We can apply statistics in a variety of situations such as the military, in organizations, and simple tasks such as household shopping. This paper aims to show the basics of statistics learned throughout the course by examining descriptive and inferential statistics. The focus will be on development and testing of hypotheses, selecting appropriate statistical tests, and evaluating results.

Descriptive statistics

Descriptive statistics is concerned with analyzing and describing data to develop configurations from it. It explains what goes on within the process of data collection. However, descriptive statistics do not allow researchers to make assumptions outside the data that the researcher has analyzed, or to make any conclusions about the previously stated hypotheses. According to Laerd Statistcis (2013), descriptive statistics are essential because if we present raw data, it would be difficult to understand and visualize what the data is representing if it is a lot. When the amount of evidence is broad, descriptive statistics helps to summarize it and then assimilate it into less complicated information. Descriptive statistics are vital because they enable us to visualize data more explicitly, thus leading to better understanding. Descriptive statistics employ two methods to analyze the amount of variety in the sets of data. These methods are Measures of central tendency and measures of variability (Tanner & Youssef-Morgan, 2013). Researchers can describe the data using figures such as graphs or pie charts to give a compelling visual representation of the data. Both of these options allow investigators to examine the number of participants in each category. For example, an investigator can create a pie chart to show the ration of men to women in a sample of the study participants and use a bar graph to indicate the number of people who exercise as always, often, sometimes, or never. Apart from using figures to present descriptive data, researchers can also use mathematical statistics to provide empirical results. The disadvantage of descriptive statistics r is that it is quite limited. It is because investigators can only make assumptions about the participants that the researcher measured during the process of collecting data. Descriptive data cannot be used to create a generalization.

Inferential statistics

Inferential statistics allow a researcher to make an accurate generalization about a specific population under study using the characteristics of the collected samples. The aspect of being able to generalize is vital because then, the results of the survey can replicate to the population as a whole. The Chi-Square, ANOVA, and T-tests are some of the methods used to determine if a specific sample is relevant to the study. Some of the inference statistics that commonly used are a regression, confidence intervals, and correlation. Laerd Statistics (2013) urges investigators to remember that inferential statistics based on using measured values in a sample to make estimates of the costs the researcher will measure in a population. However, there will always be a level of uncertainty in doing this process (Laerd, 2013). There are three essential questions that a researcher needs to ask when trying to select the most appropriate inferential statistics for study. These questions are: “What is the research question? What is the study design? What is the level of measurement?” Researchers need to consider these questions carefully when attempting to create an analysis plan or to develop a study protocol. The figures that represent these questions are meant to visually help researchers to narrow down the list of inferential statistics that would be appropriate for the study. However, the disadvantage of inferential statistics is that the data collected on a specific population that is not entirely measured cannot reliably confirm that the calculated are accurate.

Hypothesis development and testing

Testing of a hypothesis is a systematic technique used for data evaluation and decision making. It involves several steps including null hypothesis development, selecting a reliable statistical test, defining the significance of the study, data collection, and calculations, and deciding to either reject or accept the null hypothesis.

Testing of hypotheses can show associations as well as differences. A hypothesis is different from research questions because it makes a more specific prediction (Mayo and Simmering, 2015). Developing a hypothesis begins with making analytical statements about the relationship that exists between variables, while at the same time predicting the potential outcomes generated from the data tested. Many advantages come from developing a hypothesis. It helps a researcher determine the development of the research that he is conducting; it allows investigators to state the purpose of the study clearly; it enables researchers to decide upon the variable that they will measure and those that they will not in the survey; and allows a researcher to have a reliable explanation about the importance of variables.

Selection of appropriate statistical tests

During the selection of a proper statistical test, an investigator has first to explain the measurement level of the variables included in the analysis. The variables could be the ratio level, rank-ordered, or nominal. It is a step that an investigator has to accomplish for the dependent and independent variables (Simpson, 2015). The next step is to simplify the exact thing that the researcher is trying to determine with the statistical testing to select the correct analysis for the statistics. For relationships with ration-level variables or ordinal-level variables, the most appropriate statistical analysis method would be Pearson or Spearman correlations. For relationship questions having two categories of variables, the Chi-Square is the most appropriate test to use (Simpson, 2015). The last step would be to take the sample size population, calculate it, then directly relate the solutions to the statistical test that the researcher chose.

Evaluating statistical results

After completing the data testing process, the researcher must then assess the findings. Evaluating the results will help show the researcher whether or not they achieved their objective. The researcher can then take the results and decide on the kind of conclusions that can be made from the research study and data analysis. If some questions remain unanswered, the researcher can choose to analyze the data one more time or start an entirely new research process (Ott and Longnecker, 2015). The importance of evaluating the data and the results is because it enables investigators to determine whether the achieved results differed from the expected results. Overall, this process allows the researchers and or the readers to learn whether or not the data collected is reliable and also the overall success of the study.

Conclusion

In conclusion, there are two critical statistical tests that we can use to analyze the data that we have collected. We use descriptive statistics to describe what is happening with our data, while we utilize inferential statistics to make an inference from the gathered data to more generalized situations. Being able to understand and apply basic statistics to our studies makes it easy to know when trying to break down complex information in to simple variables. It is important to remember that we collect various types of data, and therefore we need to choose the statistical test that will produce the most accurate and reliable results.

References 