# Statistical Lessons

## Introduction

The role of statistics in not only daily life but also business, healthcare and governance cannot be ignored. From a general point of view, it is crucial to study statistics due to the various decisions that one makes in daily life encounters. More often, an individual might not be aware of the imperative role statistics played by study. It is evident that it infuses the majority of the decisions made each day by individuals, organizations as well as governments, both complex and simple ones. The vastness of statistics corresponds to its widespread applications in various areas. By definition, statistics refers to the analysis, construal, staging, and classification of data (Hawkins, Jolliffe, & Glickman, 2014). Data, in its entirety, does not make sense without being analyzed. Virtually, almost all human beings, whether literate or illiterate, has some comprehension of the tenets of statists. Attending statistics class only serves to improve this understanding to widen its application for professional or personal growth. Against this background, the current paper presents the crucial lessons acquired from attending statistics class. This statistics course provided valuable lessons applicable to both personal and professional life.

### Important Lessons

**Descriptive Statistics**

Firstly, descriptive statistics have significantly enhanced numerical comprehension of phenomenon. It is perhaps the most commonly applied branch of statistics. According to Harris (2014), as a branch, descriptive statistics focus on describing, summarizing, or showing data in a meaningful way to unearth simple but helpful patterns. Nonetheless, these domain of statistics does not aid much in making conclusions beyond the data that is being assessed. For this reason, it cannot be used in proving or disproving hypothesis as it will be shown in the case of inferential statistics. Despite this limitation, one cannot entirely disregard the crucial role played by descriptive statistics.

Essentially, descriptive statistics are extremely significant because they make it possible to visualize what the data is showing, particularly if the volume is immense (Lomax & Hahs-Vaughn, 2013). They offer the fastest way of analyzing the data. Based on their ability to rapidly give visualizations, one can use them to quickly determine the degree of various things in life. For instance, it is the mean, which is a measure of central tendency, is often used by people to give the normal occurrence of an event in life. Evidently, the knowledge concerning the mean and median helps in judging situations and determining the action to take when confronted with challenging environment or situations. For instance, one can choose to travel by road, air or water based on the average number of accidents occurring in such transportation media (Lomax & Hahs-Vaughn, 2013).

Additionally, descriptive statistics have significantly assisted me in coming with generalizations concerning a small-sized populations. A population refers to a group of data in which an individual interested. In essence, the size of the population can be large or small. For instance, if one is interested in students’ marks in a statistics class, then the population would be all students. In such situations, it is possible to compute the average, standard deviation and median. The mean and standard deviation will act as parameters for describing a situation(Hawkins, Jolliffe, & Glickman, 2014). Likewise, in daily life, people assign parameters to situations, which influence the decision they make.

### Inferential Statistics

Whereas descriptive statistics are crucial in developing generalization small populations, inferential statistics are necessary in studying large populaces. Inferential statistics is another important branch of statistics (Kaushik & Mathur, 2014). Nonetheless, one might not have access to an entire population in which he or she has interests. In scenarios where one can only access handful amount of data, inferential statistics are appropriate. For instance, one can be interested in the children from single-parent households worldwide. Evidently, it is not feasible to gain access to such population. The best way of approaching such problems is to select a sample to study using statistical tools and techniques. Likewise, in personal or professional life, one is always faced with bigger and complex problems that cannot be comprehended easily based on the general knowledge (Lomax & Hahs-Vaughn, 2013).

Consequently, the knowledge concerning inferential statistics is convenient in dealing with problems that are complex. With a small sample sizes, one can compute the mean or standard deviation, which are generalized over the large population (Lomax & Hahs-Vaughn, 2013). Unlike in the case of descriptive statistics, the mean and standard deviation are not referred to as parameters. In inferential statistics, they are called statistics. Consequently, it is justifiable to define inferential statistics as the approaches that facilitate a statistician to make generalizations concerning large population that contains the sample. Based on this definition, the decisions made by an individual or organization are frequently based on their ability to make generalize about the population.

### Hypothesis Testing and Development

The course also offer crucial insights about hypothesis testing and development. Hypothesis refers to a tentative extrapolation concerning the nature of association between at least two factors or variables. In other words, it is a claim made by an individual concerning the things that influence a certain situation, phenomenon or event (Lomax & Hahs-Vaughn, 2013). An example of hypothesis would be: all male students perform better than their female counterparts in statistics course. In this hypothesis, the gender of a student is the factor or variable the influences performance. This hypothesis is tentative prediction because there is no research that has proved such claim. Besides, even if research had proved this claim, the findings cannot be validated. In professional or personal life, one is frequently with new challenges that require solutions or explanations. Before tackling such problems, one tends to make assumptions, which act as hypothesis.

However, it is worth pointing out that not all assumptions or hypotheses are right. Some might be misleading from the onset. As a statistician, the assumptions made might have some element of truth and possibility of falsehood. By making valid assumptions in statistics, one can learn how to judge life situations and determine the appropriate solution. There are some conditions that must be met to ensure that a hypothesis is relevant statistically (Lomax & Hahs-Vaughn, 2013). First, it should be verifiable or falsifiable. Second, they should not be moral or ethical questions. Third, they should not be too general or too specific. Fourthly, the hypothesis should be a forecast of the results. Thirdly, it should be valuable despite being regarded as false. The latter condition also implies that it should some meaning irrespective of being false.

Every hypothesis made has its counter narrative. This leads to two kinds of hypothesis: - null hypothesis and alternative (Nayak & Hazra, 2011). The former represents theory that has been claimed since it is believed to true or because it is the foundation for the argument. However, the bottom line is that it should have been proven. In the event incorrect decision is made, the null hypothesis leads to severe outcomes. In life, null hypothesis can be likened to the things or events that one often ignores and proceeds to make decision. Such decisions frequently come back to haunt an individual because of ignorance (Nayak & Hazra, 2011). A lesson can borrowed from this fact. Consequently, it is worth claiming that statistics addresses ignorance in the process of making decisions. All factors should be considered and possibilities assessed to ensure that the correct is made and that the seriousness of an outcome is minimized.

By contrast, the alternative hypothesis refers to a statement that represents the main aim of the test. It is a negation of the null hypothesis. According to ROOO, it is only realize when the null hypothesis is disproved. To a statistician, the alternative hypothesis is the desired supposition of the research. In life, the alternative hypothesis reflects the degree that a certain an event or a situation will turn out positively.

### Selection of Appropriate Statistical Tests

Another important lesson learnt from the course is the selection of statistical tests. According to Harris (2014), statistical tests refer to the ways of making quantitative decisions concern process or situation. The primary objective of these tests is to determine whether there is sufficient proof to disprove a null or alternative hypothesis. Most people involved in decision-making frequently subject their future actions to subjective or objective assessment. These assessments can be likened to the statistical tests. When making critical decisions, one is likely to assess the possibilities of adverse effects that might arise because of pursuing a certain action. Naturally, this assessment might not be as accurate as statistical tests performed by software, such as SPSS or STAT because human beings will always have some degree of subjectivity. However, having statistical knowledge can assist an individual in avoiding biases.

Nayak & Hazra (2011) outlined five important factors that should be considered in selecting the appropriate statistical test. These variables include variables, the level of measurement of the variables, the nature of observations, nature of the hypothesis, and population or sample. Regarding variables, Nayak & Hazra (2011) claimed that bivariate statistical tests are appropriate when only one independent variable covary with the dependent one. By contrast, if they at least two, then the multivariate statistical tests are necessary. There are three measurement levels that determine the choice of statistical analysis. Nayak & Hazra (2011) cited that various statistics will be necessary if the dependent factor is measured at the ratio, ordinal or nominal level. When comparing populations, it is necessary to use various techniques because populations have not error variance.

### Evaluating Statistical Results

The evaluation of statistical outcomes was also another important concept taught in statistics course. The findings of research are only helpful if they correctly evaluated and interpreted. The knowledge gained from the course was crucial in strengthening the significance of study results. By definition, the process of evaluating statistical results involves findings a basis for disproving or proving a hypothesis (Kaushik & Mathur, 2014). It is a good practice to align the study results with the objectives. In this case, it is easier to evaluate the results against the expectations and objectives.

Once the data has been evaluated and statistics calculated, one can make appropriate conclusions. The process of analysing the results can be sophisticated based on the nature of questions the study aims to answer (Kaushik & Mathur, 2014). Despite the evaluations offering the grounds for explaining the occurrences, there might other potentially useful reasons for other events did not take place. For this reason, it is imperative not to regard the phenomenon under investigation as an isolated matter. Most importantly, some questions that might be helpful in avoiding making errors in the interpretation include: what other explanations might be available? Are the results backed by statistics? Are the conclusions rational? Are the findings different from the earlier expectations?

### Conclusion

This statistics course has provided important lessons that can be applied when making professional and business decisions. From the start, this discussion has primarily focused and reflected on the lessons learnt from statistics course. The first lesson pertain the branches of statistics, which include inferential and descriptive statistics. These branches differ significantly based on their area of application. Whereas descriptive statistics are applied in situations where the entire population is accessible, inferential statistics are crucial in making generalizations about a population that is not within reach. The second lesson learnt was the development and testing of hypothesis. Regarding this lesson, there are two types of hypotheses, which include the null and alternative hypotheses. Thirdly, the course offered important insights about the selection of appropriate statistical tests. Through knowledge obtained from the statistics course, complex problems can be analysed and resolved in a rational and objective manner.