Multiple Regression

Scenario

            I am working in a hospital and interested in ascertaining the predictors variables for flu & how severe it will be based on these variables to predict the level of severity. In this research study, we also determine the variables that predicts outcome of other relative variables such as “disease”, “academic performance”, or the “professional analysis”. This is established through determining the level of correlation of the variables with the response variable in correspondence. These predictions on the outcome are thus formed through examining how well the predictor variables in the study correlates with the response variable (Mosteller, & Tukey, 1977). Now, we ascertain the potential variables that are to be examined in terms of the severity of the experience of the individuals with flu.

Predictor Variables for flu severity

            The possible variables that may predict severity of the flu are, age, exposure to pollutants, medical conditions, stress, exercise, seasonal incidence, climate, and travel. These variables are potential predictors that can predict the severity of the flu.

            The age of the individual will be taken into consideration because the age is a determining factor of the disease catching age. Previous medical conditions related to level of severity of flu. The exposure to pollutants increases risk of the flu and higher it is higher the severity will be. Stress factors contributes to severity of flu, level of exercise is a relative factor, seasonal incidence is a relatively doubtful factor taken into consideration for testing the relativity of the factor in the research study in correspondence, travel & climate factors also relate to the impact of level of flu in an individual.

Measuring each variable

            There are four scales of measurement in terms of which the variables will be measured. We measure the variables under different scale and it will be as follows;

Variable

Scale of Measurement

Age

Ordinal

Exposure to pollutants

Nominal

Medical Conditions

Nominal

Stress

Nominal

Exercise

Ordinal

Seasonal incidence

Scale

Climate

Ordinal

Travel

Nominal

Table showing variable measurement level

Type of multiple regression analysis: Standard, Sequential, or Stepwise

            The multiple regression that will be used is standard regression. We run the test to ascertain the level of severity of flu based on the eight respective variables. The standard regression would be better to ascertain the linear relationship between the variables. We make assumptions on the predictor variables as whether it would change the level of change in the response variable. The test is pat since we want to predict the change response variable based on chosen predictor variables (Aiken, West, & Reno, 1991). Neither do we not have any prespecified criterion hence not facilitating use of stepwise regression. Further we do not have large set of predictors. Not do we not have any order of entering the predictor variables based on any order.

References

Aiken, L. S., West, S. G., & Reno, R. R. (1991). Standard Multiple Regression Analysis: Testing and interpreting interactions. Sage.

Mosteller, F., & Tukey, J. W. (1977). Data analysis and regression: a second course in statistics. Addison-Wesley Series in Behavioral Science: Quantitative Methods.

In multiple regression, the predictor variables can be measured on any scale of measurement but the outcome is always measured on a continuous scale. What are the different things you might want to interpret in your analysis? In a multiple regression we can evaluate both the significance of the overall model as well as the significance of each predictor. What might you expect with each in your example?

Cite any sources in APA format.