Pre-Registration Analysis Plan, SWOT Analysis Examples

Statistical Models

Various statistical models are employed to analyze various sorts of data. There are different statistical models because different variables are used in researching different aspects. The three types of models are parametric. Semiparametric and non-parametric models. The statistical model used in the analysis of data in their case is Analysis of Variance (ANOVA). The tool is essential since it splits the observed variability in a dataset into two segments; systematic and random factors. The analysis will involve determining the cause-and-effect variable, whereby the variable causing the other action must be determined from the analysis to be effective.

This research aims to determine the reaction of participants who are not factored in and hence do not qualify for an interview they have attended. One of the participants who is also a contestant for the job is favored by the policy actions in this case. Therefore, there is a need to use the variables provided in the codebook to determine how the other participants felt about the issue. Different variables determine what reactions will be triggered in the participants. Such variables include the age of the participants, gender, work experience, and other work experience, which affect how much people will react. Gender, in most cases, determines people’s reactions because males and females react differently to the same issues. The age or work experience might also make a person react differently to the same issue as another person reacts hence a major determining factor.

The data provided in the codebook is enough as the aspects provided in the codebook are different; hence, little or no deviations will be witnessed from the results obtained from the research. Multivariate ANOVA (MANOVA), a two-way method of testing the concepts, will be applied in testing the aspects in this case (Onana et al., 2020). The method will be used since there seem to be two dependents. Therefore, there is a need to ensure that all the aspects or characteristics that cause another thing to happen are captured correctly since many disconnections can arise when one method is used in such a case. Disconnects arise when the data provided in one case phase does not add up to the same thing as the outcome, or the result of the research differs from the other factors that contribute to the same outcome.

The values used to attach the variables to the experiment’s outcome are not deviating from the main cause with large deviations. When values used in research do not deviate so much from each other, the research outcome is similar, so the research was done best. The missing values are also minimal. The minimal cases of missing values indicate high levels of accuracy in the experiment. For example, the gender of the participants and the age of the participants have no missing values. When the conclusions of the experiments are made, the results will be close to the real thing on the ground. All the values taken will indicate the same thing (Azizi et al., 2022). ANOVA is also a good statistical model since it helps indicate all the variables essential in the research and helps analyze the participant’s reactions in this case.

Transformations

The reaction of the participants on the issue of favoritism does not need additional transformations. However, suppose the model used for regression analysis and different intensities and reactions of the participants categorically described, e.g., bitter, very bitter, somewhat bitter. In that case, bitterness could be dummy coded with not bitter as the reference category. Bitterness, in this case, can be correlated to being angry since the people showed their discomfort after the people who were conducting the interview decided to be unfair to them.

Different coding methods will be consulted in this case, making it possible to compare and contrast the results that will be achieved. Dummy coding and effect coding are the best methods of coding. Dummy coding enables the researcher to use categorical predictor variables in different estimator models. Using different estimator models also helps get mixed results and characteristics of different variables tested in every case. When different aspects of different predictor variables are tested, coding and making conclusions on the results or the outcome becomes easy as a person can get all the details and characteristics of different datasets. The model allows a researcher to assign weights to the various categories of categorical variables. Different research methods and data analysis, including coding, make it possible to get different aspects of the data sets that could not have been noted if the data was not coded using other means.

Inference Criteria

When analyzing a given study, the researchers must ensure that all the variables are well set and categorized in the best way to enhance the accurate drawing of conclusions based on the results obtained from the data. The data analysis on the respondents’ behavior and feelings when some of them received favors in contrast to what they anticipated according to the requirements of the hiring process can be concluded using a variety of approaches. The ANOVA and post hoc test will be evaluated using P values, the accepted criterion, which show that the outcomes differ from those expected if the null hypothesis is valid.

To avoid problems with the p values and confidence level intervals, researchers report the hypothesis values and let the people who conduct the analysis interpret and get the statistical significance by themselves. The p method of hypothesis testing is important and convenient as the errors encountered in this method are minimal. The p values also enhance transparency which is the most important thing in data analysis since the researchers can disclose all the types of data used in the tasks and the significance of all the variables used in the research. People who were not involved in the research can also get all the concepts of data and all the significance of all the information provided regarding the case at hand.

Data Exclusion

Awareness checks must be undertaken because some factors will only present challenges and errors when conducting the data analysis exercise (Neamat & Hassan et al., 2021). There is a need to ensure that the eligible data and variables safe places in a given file under a creating category that will be different from ix with the others. Some of the data used in data analysis does not add any value to the research; therefore, errors are the only thing it brings up in the analysis. The errors are caused by meaningless data, which should be there only when needed but not at all times.

In the case of this study, the factors that make people angry and bitter about the issue of favoritism of some of the people who were to be taken based on the merits they had for the job and the requirements they had fulfilled. Some of the most significant merits that should be given priority are the age and gender of the participants. The work experience and additional requirements that might boost a person’s chances of achieving the job should also be considered when deciding how the other participants will react. Some factors include the time when all the participants arrived, the glooming of the participants, or their status. Such factors can only be considered by the people holding the interview but cannot affect how the participant reacts to the issue of favoritism. Bitterness is emotional and can only be triggered by the factors and variables above. No other eligibility techniques will be applied in the data analysis task.

Missing Data

In some cases, the task being undertaken might be delayed due to a lack of data that should be used to facilitate it. The researchers should ensure that all the data needed is kept in a safe place that unauthorized people cannot access and that all those who handle such data are professional to avoid data loss. Data is the main thing in this case since the analysis cannot be done without the data. Researchers must also keep in mind that all the datasets to be sued and all the formulas, including the variables essential in the data analysis, are complete to avoid any data analysis being left hanging due to the unavailability of the data to complete the task.

Missing data from variables or formulas will be treated as things that should not be included in the analysis. Excluding variables with missing data, will be convenient as no aspects will be made and left without explanations on how the conclusion arrived. It is important when one is talking about a certain aspect. They have the supporting data and evidence that is making them say so. Some things that make such people fail to have enough evidence are the lack of that data sources and the procedures they should take to make the best decision. In short, no variables with missing data be included in the data analyses so that when the analysis of the reactions of the participants is being done, enough evidence can be deduced from the provided data sources and the information available in the datasets, modes used, formers applied and the information available in the codebook.

Exploratory analysis

A test will determine the relationships between the data attained from the experiment. According to the researchers, certain things are similar to one another in some way. Therefore, when the comparison is being made, some traits based on the type of data being handled will come out themselves. When the traits are visible after the data analysis has been done, it will be easy to repeat the procedure so that the researcher can ascertain that the results obtained and the analysis done is similar and accurate.

The repeated procedure will also make the researcher feel comfortable as they can do it faster and link up some of the things that will come up in the experiments. In these cases, some things that might be similar in one way or another are the reaction of people who are more aged than others. Older people are likely to recta in a more mature way than the young ones. In most cases, the young react violently to the situation, which might cause chaos and disruption.

On the other hand, when it comes to gender issues, women are deemed to be softer than men and, therefore, will react less violently compared to men. Individuals who have struggled and gone through the system are also likely to react more maturely compared to those who are trying for the first time since they might be frustrated and react violently. Such things are the ones that are said to be linked and intertwined since the participants, in this case, come from the same environment and share some common things.

References

Azizi, F., Ghasemi, R., & Ardalan, M. (2022). Two Common Mistakes in Applying ANOVA Test: Guide for Biological Researchers.

Neamat, S., & Hassan, M. (2021). A Review on Using ANOVA and RSM Modelling in The Glass Powder Replacement of The Concrete Ingredients. Journal of Applied Science and Technology Trends, 2(02), 72-77.

Onana, A. B., Njikeu, O., & Onana, R. M. (2020). Cameroon found parameters that provide higher boreholes flow rates using the ANOVA test. Sustainable Water Resources Management, 6(5), 1-14.