Statistics for Research Methods II

Statistics for Research Methods II

Binary Logistic Regression

According to Strickland (2017), a binary logistic regression is a unique type of regression in which the binary experimental variable is linked to a set of independent variables. The independent variable can be interval or ordinal, discrete or continuous, numerical or categorical.

1. Under what circumstance is a binary logistic regression used?

1. When there is only one single outcome
2. When there are outliers
3. When the outcome variable is dichotomous.
4. None of the above

Ordinal Logistic Regression

Ordinal logistic regression is a special form of regression. As Gero (2016) explains, an ordinal logistic regression is employed in forecasting an ordinal response variable given one or more explanatory variables. Moreover, the model also uses interactions between explanatory variables to project the dependent variable.

1. Which of the following statements is not an assumption of the ordinal logistic regression?

1. Ordinary logistic regression assumes that the response variable is ordinal.
2. Ordinal logistic regression presumes that the influence of the explanatory variables is similar for all values of the predicted
3. Ordinal logistic regression assumes that there is no near perfect linear relationship among the predicting variables.
4. Ordinal logistic regression assumes that explanatory variables are not continuous.

Odds Ratio

The odds ratio is a statistical tool used in quantifying the strength of the relation between two measures and is regarded an effective instrument for discerning relationships between categorical variables (Rautaray, Eichler, Erfurth, & Fahrnberger, 2020). The tool illustrates the ratio of the likelihood of occurrence A in the presence of a second event, B, and the probability of incident A in the absence of occurrence B.

1. Which of the following statements is true?

1. An odds ratio informs an individual the degree in which the odds of the response variable change for each unit change.
2. An odds ratio, which is less than 1 implies that the odds reduce as the explanatory variables decrease.
3. An odds ratio equal to 1 indicates that the odds change as the explanatory variable elevates.
4. An odds ratio, which is greater than 1 implies that the odds change as the explanatory variable heightens.

Goodness-of-fit

In particular, the goodness-of-fit is a widely used statistical measure. According to Asante-Duah (2017), the goodness-of-fit measures the degree in which an empirical distribution fits the distribution anticipated under a normal distribution. The chi-square goodness-of-fit test is employed in determining the magnitude of the difference between the distribution suggested by a data sample and a pre-selected probability distribution (McCuen, 2016). The chi-square test stands as one of the broadly used one-sample analysis for evaluating a population distribution.

1. Which of the following tests is not an example of a goodness-of-fit test?

1. T-test
2. Chi-square test
3. Shapiro-Wilks’ test
4. Kolmogorov-Smirnov test

Odds

In particular, odds are described as the likelihood that an event will take place divided by the probability that a particular event will fail to take place (Boston University School of Public Health, 2017). For instance, if the likelihood of a particular occurrence to take place is Y, it follows that the prospect of the event not to take place is 1-Y. The odds of an event illustrate the ratio of the possibility of an incident to happen against the likelihood that an occurrence will not take place.

1. If the probability of Germany winning the football world cup of 2022 is 0.1, what are the odds of it not winning the competition?

1. 6
2. 19
3. 4
4. 9