What I Have Learned About Statistics

The complexities occurring in all aspects of the world make decision making necessary to make good sense of what is really happening. Mathematics offers the most accurate means of deciphering meanings into scientific complexities. Statistics entails the science oriented collection and analysis of massive information or data with the purpose of deducing meaning of outcomes generated from a representative sample (Ang & Van Dyne, 2015). It involves the application of diverse scientific techniques for collecting, organizing, analyzing and reading data towards describing and making sound decisions. Statistics is universally relevant especially in cases where one seeks to further develop educational standards to encompass research studies. This research paper highlights my understanding of descriptive statistics; inferential statistics; hypothesis development and testing; selection of appropriate statistical tests; and evaluating statistical results as learned through the semester.

Descriptive Statistics

Statistical numbers and averages regularly quoted concerning financial performance of stock exchanges in different parts of the globe are essentially descriptive statistical figures. Descriptive statistics attempts to encompass huge observable data sets towards offering audiences with a generalized idea concerning the population (Cressie, 2015). Central tendency measures such as the mode, median and mean; data distributions such as normal distribution as well as standard deviations are used to provide basic descriptive statistics. For instance, in gaining demographic information about a particular community can be termed as a sample size. Descriptors such as gender, age, and financial exposure are dependent on what information the researcher seeks to achieve relative to an understanding of the experiment. Descriptive statistics assists researcher in plotting simple histograms to gain basic understandings concerning occurrences with a given experiment.

Inferential Statistics

Inferential statistics is normally applied in making sense of conclusions drawn from a particular experimental endeavor with the aim of acquiring more answers to questions about samples and populations not addressed in the specific experiment (Griffith, van den Heuvel & Fortier, 2013). For instance, a researcher seeking to ascertain claims that a particular training module enhances test scores usually begins with two groups. One group does not take the module and is referred to as the control group while the other involves active study participants. Measuring test scores prior to and at the end of the experiment ensures certainty that starting scores are based on an average encompassing both groups. If data from the experiment shows that the control group performs lower than the study participant and that results are statistically significant, then the claim is true.

Hypothesis Development and Testing

In most cases, research is motivated by a problem. The hypothesis is essentially a query generated by experimenters during speculation on the outcomes of a study. True experimental design regards such a statement as critical to its structure given that it highlights the ultimate aim of any given experiment (Harlow, Mulaik, & Steiger, 2016). The hypothesis development process begins with the isolation of a research question. This informs the research topic which begins from a varied area of interest to a critically defined issue. It is dependent on researcher skills towards ensuring the requirements of the entire study are fully catered for. Good hypothesis development ensures logical consistency, simple and clear statements, relevant to current literature, and testable.

Hypothesis testing begins immediately after development through the isolation of independent and dependent variables (Harlow, Mulaik, & Steiger, 2016). The valuable dependent’s values which are often predictable from the independent variable provide a basis for hypothesis testing ensuring that the independent variable is presumed as the cause of the study.

Selection of Appropriate Statistical Tests

Employing wrong statistical testing ultimately results in an unworthy research outcome. This implies that appropriate statistical testing is a preeminent procedure in research data analysis. It involves comprehensive understanding of research study variables in line with study objectives and whether a selected test investigates data based on type.

Evaluation of Statistical Results

After data collection and the subsequent analysis process, it is imperative that the validity of acquired statistical data is evaluated. Assessment of statistical results further enables for the interpretation and analysis of numerical or categorical data (Harlow, Mulaik, & Steiger, 2016). This requires the pooling together of all pertinent data for a particular sample based on specific statistical parameters. Standard deviation as well as the mean can then be calculates to allow procedures to continue on the determination on the p-value of the selected data range. Evaluation of statistical results allows for this value that is concerned with highlighting differences between the hypothesis, null hypothesis, and alternative hypothesis as formulated at the beginning of an experiment (Harlow, Mulaik, & Steiger, 2016). The p-value enables the experimenter to determine whether there is a valid null hypothesis or not. It is upon the researcher to appropriately select a statistical tool which adequately facilitates for a sound evaluation process. The statistical tool used is however dependent on the form of statistical data requiring evaluation.

Conclusion

Very many aspects of our world are becoming more dependent on big data. Statistics is critical to enabling persons acquire fundamental knowledge and skill sets necessary in making sense out the massive information available irrespective of the field of research. The knowledge gained allows for good comprehension of the statistical terminologies, applications, techniques, and definitions describing graphically represented data as well as understanding of frequency distributions. This paper has further shown the necessity of inferential statistics in using available research information to gain understanding on emerging issues or setting up of upcoming experiments. Furthermore, hypothesis development and testing have been highlighted as vital to the success of an experiment and ought to be tested through statistical tests which ascertain viability of a hypothesis.

References

Ang, S., & Van Dyne, L. (2015). Handbook of cultural intelligence. London, UK: Routledge.

Cressie, N. (2015). Statistics for spatial data. Hoboken, NJ: John Wiley & Sons.

Griffith, L., van den Heuvel, E., Fortier. I., et al. (2013). Harmonization of cognitive measures in individual participant data and aggregate data meta-analysis [Internet]. Rockville (MD): Agency for Healthcare Research and Quality (US); 2013 Mar. Methods and Results: Implementing and Evaluating Three Methods of Statistical Harmonization Applied to Cognitive Measures (Objective 3) Available from: https://www.ncbi.nlm.nih.gov/books/NBK132542/

Harlow, L. L., Mulaik, S. A., & Steiger, J. H. (Eds.). (2016). What if there were no significance tests?: classic edition. London, UK: Routledge.

  Do you need high quality Custom Essay Writing Services?