Regression
In a sense, a well-fitting line drawn through the plot of a strong correlation gives us a tool for predicting the performance of a variable on one measure if we know it's score on the other. Informally, most humans do regression, which is the process of predicting future outcomes based on past performance. Let's say your scores on the past 3 tests in a class have been 75, 82 and 85. What prediction would you likely make about your performance on the last test? Figuring this is the process of regression.
You can use regression analysis for the following things:
- Finding correlations
- Analyzing study results
- Predictinf future outcomes
In order to do this formally, in the statistical arena, we need to make an effort to assess the dependability or reliability of the past as a guide to the future. We will use some of the tools we've acquired to this point, and learn a few new ones. Sometimes when we try to do this prediction, we use the mean. Let's say that we are planning a trip to Rome, Italy in September. We need to pick a suitable wardrobe. We do some research and discover that the mean daytime temperature in Rome at that time of year is 75., so that is our best prediction. The mean nighttime temperature is 45 so now we will be appropriately prepared for our time outdoors, day or night.
In many situations we have a lot more information than this to work from. We may have an entire distribution of observations, so we can figure the mean and standard deviation. If we are looking at our own score in relation to the others, we can interpret the meaning of its difference from the mean by doing a transformation and using z-scores.
However, as much as we cans speculate on future performance based on the above-mentioned statistics, they really aren't sufficient for predictions that we can have some confidence about. We need to use our recently gained correlation skills to help us understand the relationship between the two variables.
Related Research Paper Topic Suggestions
Action Research - Action research is a form of self-reflective research paper. Custom written action research papers from Paper Masters for Undergraduate or Graduate level students.
Collaborative Research - Paper Masters specializes in collaborative research and theory driven research papers. Paper Masters can help with your collaborative research project or find you refrences that are examples of collaborative research in any field of study.
Empirical Article - Empirical Article research papers examine a type of writing for articles that relies on scientific data derived from observation and/or experimentation.
Evaluation Research - Evaluation research from Paper Masters provides custom written applied research that is based on systematically evaluating a program, action or policy.
Experimental Method - Experimental Method is a type of research methodology that uses either quasi-experimental or true experimental methodology to conduct a study.
Historical Comparative Research - Custom written historical comparitive research projects from Paper Masters - Specialists in all forms of research studies at the graduate and undergraduate level.
Inferential Statistical Procedures - Analysis of inferential statistical procedures in academic research papers and studies. Capstone projects, Theses and doctoral projects often include inferential statistical procedures.
Literature Review Approach - How to approach writing a literature review using top-down and bottom-up research strategies.
Textual Criticism - A research paper on textual criticism look into the study of variations of different texts from different time periods.
Theoretical Inquiry Research - Theoretical Inquiry in research papers from Paper Masters match your exact needs because they are custom written by professional writers.