Introduction
In social research methods, inferential statistics has a core role to play in reaching the conclusion of the analysis. There are two types of analysis involved in a research study inferential statistics and the descriptive statistics. The amount of data gathered from different subject groups cannot solely help deduce the outcome and support of the hypothesis. Inferential statistics facilitates the researcher come up with a conclusion that goes beyond the immediate data. On the other hand, descriptive statistics enable the researcher to deduce information from the given group of data. This case study contributes to explain the various tests used in inferential statistics and their appropriateness.
Body
Inferential statistics surmise information from the sample to the general population. They are important in determining the strength of the relation between the independent and dependent variables in a model. The use of inferential statistics has helped in making policy recommendations because the analysis provides the empirical support. The researchers make use inferential statistics in social research studies under the following conditions. First, the data present should comprise of a full population member (Denzin & Lincoln, 2000). Secondly, a random sample is drawn from the entire population. Finally, through the use of a predetermined formula, assess whether the sample size is large enough. The following are some inferential statistics used in social research method.
One sample test of difference/ one sample hypothesis test
When is it used?
This test is the simplest inferential tests that are utilized when comparing the average performance of two clusters on a single measure to establish a difference (Loether & McTavish, 1993). The t-test analysis is suitable when comparing the means of two study groups.
Why is it used?
The t-test is commonly used because it is easy to use, and draws a conclusion on the variability of the two groups. The variability helps determine whether there is a statistical difference in the means of two groups.
How is it interpreted?
The significance level is tested to establish whether the ratio is big enough to conclude that the difference between the groups was certainly not a chance (Alston & Bowles, 1998).
Confidence interval
When is it used?
The confidence level is used to evaluate a score or a value in a population subject to the score of the sample participants.
How is it interpreted?
Using the level confidence method, degree of confidence is used to infer test, the test results (Asadoorian & Kantarelis, 2005). A 95% confidence level means that the researcher is 95 percent confident to conclude the score from a given sample range.
Chi-square and Contingency tables
When are they used?
These types of analysis are used in a situation where there are two categorical variables. For instance, to determine the relationship between gender and the scores on measurement outcome, these methods are utilized.
How are they interpreted?
The Chi-square method is used to establish the strength of a relationship. The relationship between gender and outcome score is analyzed. Does knowing the gender of a student predict the outcome score? Therefore, if the probability of the chi-square statistics is 0.5 or less, an informed conclusion is deduced that independent variable predicts score on the outcome variable (Burns, 2000). Contingency tables are used equally for comparison across the independent variable on the outcome or dependent variable (Asadoorian & Kantarelis, 2005). For instance, compare the proportion of male students who agree on a certain idea to the proportion of female students who agree to the same.
T-test/ ANOVA
When is it used?
The T-test or the ANOVA statistical method is used when there is a categorical and continuous variable. Comparison of mean scores between two or more groups is done using the ANOVA analysis (Hek, Judd, & Moule, 2002). For instance, compare the mean of a group of students lectured across race.
How is the analysis interpreted?
The ANOVA analysis of T-test is used to establish any significant difference between the group means. If the probability attributed to the F statistics is 0.05 or less, the researcher can conclude that there is a difference in the means.
Pearson correlation
When is it used?
The Pearson correlation technique is used to explore the relationship between two quantitative, continuous variables (Hek, Judd, & Moule, 2002). It is also referred as the correlation coefficient. This technique measures the strength of association between two variables, for instance, age and blood pressure.
How is it interpreted?
If the probability attributed to the T statistics is 0.05 or less, then the assumption is made that there is a relationship between the independent and dependent variables (Loether & McTavish, 1993).
Steps involved in creating a model
There are two types of research models, the quantitative and qualitative paradigms. A model is explained as a specific strategy designed to address an education research problem. A quantitative paradigm is usually a deductive process while the qualitative paradigm is an inductive process. When developing a research method; it is advisable to work on an earlier theoretical model. Research models are meant to address the research problem and infer the research objectives. The initial step in the creation of a model/ paradigm is identifying an idea. In research methods and methodologies, an idea is described as a concept. An idea is structured in a manner that it is interesting to draw attention to the populace before even establishing its worthiness.
Originality is an important aspect when creating a model as it avoids overlooking at the literature for alternative paradigms. The entire procedure has myriad challenges and complexities as it involves a lot of refinement. It is time-consuming and entails a step by step verification of the ultimate model. Various paradigms are available in inferential statistics that facilitate the analysis process. Nevertheless, care should be observed when choosing either of these models as it could fabricate the conclusions.
Describe two different studies of interest
A study on the impacts of Brexit strategy on the European economic growth and development
Research questions
How does an exit strategy affect the Britain pound?
Does the exit strategy influence other economic parameters?
What measures and policies that can help alleviate these impacts?
Research hypotheses
The null hypothesis of this study is:
Ho: As a result of Brexit strategy, there will be no significant difference in the economic growth and development in Britain.
Tested against the alternative hypothesis:
HA: As a result of the Brexit strategy, there will be a significant impact on the Britain's economic growth and development.
Variables
In the case of study, the variables to be put under scrutiny are economic growth, the Brexit policy, balance of payments, currency, and inflation rate (MacShane, 2015).
Another study of interest is the correlation between female genital mutilation and academic performance in developing economies.
Research questions
What are the factors that lead to female genital mutilations?
What are the effects posed by female genital mutilation?
How can the female genital mutilation be stopped?
Research hypotheses
The null hypothesis of this study is:
Ho: There is no significant correlation between female genital mutilation and academic performance of the girl child
Tested against the alternative hypothesis:
HA: There is a significant correlation between female genital mutilation and academic performance of the girl child
Variables
The variables present in this study include age, religion, and knowledge. In the female genital mutilation, age has a vital role to play. According to research studies, mutilation is performed on young girls between the tender ages of 10-15 years. Various religious groups have different perceptions on the female genital mutilation with most of them opposing the evil deed. Sound knowledge of the practice among the populace plays a positive role in curbing FGM (Kaplan, Hechavarría, Bernal, & Bonhoure, 2013).
Advantages and disadvantages of statistical models
Statistical models help provide detailed information compared to descriptive statistics, giving an insight into the relationship between the research variable. Research studies have been carried out in business and academics, providing a convincing support of the given theory.
Limitations of Statistical Models
In analyzing data, the researcher uses unverified information about the population putting the correctness of the values into question. Inferential statistics is pegged on the concept of using scores measured in a sample in making a conclusion. As a result, a certain proportion of uncertainty will occur in undertaking the process. Some inferential tests require the researcher to make predictions and educated guesses. Additionally, this will influence the certainty of the results to some extent.
Inferential statistics has been used in various research studies to come up with recommended and suitable policies and regulations concerning a particular research problem. Social research methods rely on inferential statistics to infer information from a given descriptive statistics. These models are useful in the future as they provide the basis of a research model and specification for various research methods.
References
Alston, M., & Bowles, W. (1998). Research for social workers. An introduction to methods. Australia: Allen and Unwin.
Asadoorian, M. O., & Kantarelis, D. (2005). Essentials of Inferential Statistics (4th ed). Maryland: University Press of America.
Black, K. (2012). Business Statistics: For Contemporary Decision Making. New York: John Wiley & Sons.
Burns, R. (2000). Introduction to research methods. London: Sage.
Denzin, N., & Lincoln, Y. (2000). Handbook of qualitative research (2nd ed). London: Sage.
Hek, G., Judd, M., & Moule, P. (2002). Making sense of research. An introduction for health and social health care practitioners (2nd ed). London: Continuum.
Hoyle, R. H. (2015). Handbook of Structural Equation Modeling. London: Guilford Publications.
Kaplan, A., Hechavarría, S., Bernal, M., & Bonhoure, I. (2013). Female Genital Mutilation/Cutting: The Secret World of Women as Seen by Men. BMC Public Health, 1-12.
Loether, H. J., & McTavish, D. G. (1993). Descriptive and Inferential Statistics: An Introduction. New York: Allyn and Bacon.
MacShane, D. (2015). Brexit: How Britain Will Leave Europe. London: I.B.Tauris.
Sahu, P. K., Pal, S. R., & Kumar, A. (2015). Estimation and Inferential Statistics. New York: Springer.
