The process of data analysis is critical in drawing valid observations from collected data. Data analysis is a major step in any research as it is the process through which, one can draw valid observation from collected data. During the class, I learnt that it is only through data analysis that a researcher can convert collected data into information. Without the process of data analysis, the data collected may not make sense and therefore, one cannot make any meaningful observations from the data. I have also learnt about the various methods that one can use to analyze data. Some of these methods include the use of descriptive statistics, inferential statistics, evaluating statistical results among others.

First, I have learnt the importance of descriptive statistics in data analysis since they help the data analyst in describing or summarizing data in a meaningful way. Through the use of descriptive statistics, one can recognize any patterns that may emerge from the collected data (Bickel, & Lehmann, 2012). A limitation of all methods of descriptive statistics is that they do not allow the analyst to draw conclusions that extend beyond the sample or population. The descriptive statistics only represent the selected sample. Moreover, one cannot test a hypothesis using descriptive methods. One can classify descriptive statistics into either measures of central tendency or measures of spread. Measures of central tendency include the mean, mode and median and they describe the central position in a sample. On the other hand, the measures of spread are useful in describing the distribution of the statistics in the sample, and they include the variance, standard deviation, interquartile range, quartiles and the range.

Regarding inferential statistics, they are methods that allow one to draw conclusions about an entire population from a sample taken from the population (Lowry, 2014). Therefore, the inferential statistics treat the collected data sample as being representative of the entire population. There are two methods involved in inferential statistics, that is, statistical hypothesis testing and parameter estimation. Therefore, through use of inferential statistics, one estimates the results that one would derive while using parameters. However, in the process of using inferential statistics, errors naturally occur. It is not possible to select a sample that represents the population accurately. To minimize the errors, it is important that one should select a sample that is representative of the population. One can select such a sample through randomization.

Hypothesis developing is one of the most important steps in the initial stages of a research. The hypothesis developed determines the kind of research that one will conduct. Moreover, the hypothesis determines the types of analyses methods one will apply. Additionally, the hypothesis development also allows one to structure the research such that the research design meets the aims of the research. Hypothesis testing is also an essential part of the research process. It determines whether the research was successful or not. There are various methods that one can use in hypothesis testing. Once one has completed the hypothesis testing, one can either reject or fail to reject the hypothesis. If one fails to reject a given hypothesis, then, the hypothesis that one developed is true.

I have also become familiar with selection of a statistical test, and that it is a difficult process. Indeed, selection of a wrong test would lead to incorrect conclusions from the data (Nayak, & Hazra, 2011). There are various factors that affect the type of tests that one would conduct. For instance, one should determine the level of measurement that one needs to include in the analyses. Data levels include intervals, ratios, ordinals and nominal. One should determine data levels for the dependent variables as well as the independent variable. The second factor affecting the type of statistical test conducted is the objectives of the study. For example, the objectives of the study can determine whether one should use ANOVA or MANOVA during analysis. Similarly, ordinal and ratio data level requires one to use Spearman and Pearson Correlations in data analysis. Lastly, one should also consider the sample size used. For instance, the sample size determines whether one should use a t-test, chi-square test, correlation or linear regression.

Evaluating of statistical results allows one to draw conclusions from the analysis conducted on the data. In evaluating statistical results, one should identify the issues in the research that one requires their information. One can determine these issues through evaluating the objectives and the hypothesis. Secondly, one should determine the appropriates analytical techniques for the sample. For instance, one can use graphical methods such as tables and graphs. Thirdly, one should assess the results of the analysis to determine whether they meet the objectives set in the study. The evaluation process allows one to draw appropriate conclusions from the data and to make recommendations.

In conclusion, statistics are important in conducting a research. It is only through analyzing statistics from a sample that a researcher can make observations from a sample. Moreover, statistics also allow a researcher to estimate population parameters from a given sample.