Description
RES6003-Module 8 AssignmentCorrelational Tests in Microsoft ExcelCorrelational procedures find relationships between two variables, rather than difference. Causation, however, must be applied carefully to this model.In this assignment, you will conduct a correlational test in Microsoft Excel.Step 1. PrepareOpen the Sample Dataset you created in Module 4 and are using for the remainder of the course. For this part of the assignment, the following null hypothesis will be tested:H0: There is no statistically significant relationship between mean project grades and mean final grades.This says the Pearson’s correlational coefficient r will be 0.Step 2. TestTo test this hypothesis in Microsoft Excel, select the “Data Analysis” button from the Data toolbar and Analysis ribbon. A dialog will appear showing a list of available statistical tests. Choose “Correlation.” Another dialog will appear giving you options.For the Input Range, select Columns C and D on our dataset. Our data are grouped by columns, so select that option. Check “Labels in First Row” since we labeled our columns Project Grade and Final Grade. Choose “New Worksheet” for “Output options” and click “OK.”Step 3. ReviewExcel will generate a new worksheet with the results of the correlational test. Review the results, noting the following:For each relationship, Excel will generate a correlation coefficient, known as Pearson’s r, a value between -1 and +1.The closer the value is to 1 or -1, the stronger the correlation. The closer the value is to 0 in either direction, the weaker the correlation.A value closer to +1 indicates a strong positive correlation and a value closer to -1 indicates a strong negative correlation.Step 4. VisualizeTo visualize the relationship between the two variables first insert a scatterplot on the Sample Dataset worksheet:Select Columns C and D in the dataset by highlighting them with the cursor.Click on the “Insert” tab of the toolbar ribbon and select “Scatter” in the Charts group (the simplest icon with perpendicular axes and dots).Then, insert a 2D line graph with two series (Column C and Column D), project grade, and final grade:Click on the outer area of the scatterplot, then the “+” for Chart Elements.Click on “Trendline” to insert a checkmark in that box.Click on the arrow to the right of “Trendline” and select “Linear.”A dotted line will appear representing the best-fit line for the dataset. A line with a slope of 1 would be angled upward at a 45° angle; the line with a slope of less than 1 would be flatter and a line with a slope greater than 1 would be steeper.Step 5. AnnotateAnnotate a cell on the Excel worksheet noting whether the null hypothesis based on the calculated value of Pearson’s r could or could not be rejected. Does the scatterplot visually match this calculation?Step 6. SubmitSave and submit your Excel workbook.Note: You will continue to use this workbook for the Module 10 assignment.Module 8 DiscussionWhat Is the Difference Between Correlation and Regression?Correlation and regression are tightly related, but they serve slightly different purposes. A regression predicts and describes the relationship between two variables whereas a correlation only identifies if there was a relationship between them in the past.For this discussion, develop an original response describing a scenario where you might need to use regression instead of correlation. How could a regression model serve your purpose?In this open-ended discussion, designed to generate a free flow of ideas, no APA citations are required.Please read the Discussion Guidelines before posting your responseModule 8 ReadingsUsing and Interpreting Statistics in the Social, Behavioral, and Health Sciences – William E. Wagner III & Brian Joseph GillespieRequired Reading:Chapter 11: How to Interpret and Report Regression ResultsFull Text Access:Follow the steps to login via Institution (American College of Education) to the SAGE website to read eBooks:1) Click the provided reading URL. 2) In the top-right corner of the SAGE website, click the Sign in: “Institution” button, and then click “Access via your institution.” 3) In the box, begin typing American College of Education and select it from the drop-down menu. Check the box for “Remember my institution” and then click the green “Continue” button. 4) Click the green “OpenAthens/Shibboleth” button. 5) Click on “ACE Student and Alumni Login” and then enter your credentials/login as usual.After logging in you will be redirected to the eBook on the SAGE website.The Tao of Statistics: A Path to Understanding (With No Math) – Dana K. KellerRequired Readings:Chapter 37: Likely Range – Confidence IntervalsChapter 38: Association – CorrelationChapter 39: Predictions – Multiple RegressionChapter 40: Abundance – Multivariate AnalysisModule 8 supplemental lecture and videoHello Scholars,Let me start off by saying that a correlational study CANNOT EVER DETERMINE CAUSE, EFFECT, IMPACT, OR INFLUENCE. The ONLY way to determine these are using a true experimental design with random group assignment for participation such as treatment vs control groups. The reason for this is because when we cannot control for all factors affecting the study variables, we cannot ever be certain a confounding variable created a significance.Academia has recently changed this view on this and will now accept crappy science and crappy research. I do not care what they say, I will not accept any form of cause or effect generalization from any research design that does not have this random group assignment. Here’s an example:Is there a statistically significant difference between schools (elementary vs middle school) in the weekly average amount of time spent on homework?Elementary of middle school is a naturally occurring group. I do not assign a student to elementary or to middle school. No random assignment is there. Now, If I found a difference, I cannot say that the school was the cause or the effect. What if one school has a different homework policy? What if one school had parents who supervised homework time more? What if one school had teachers who offered prizes for complete and timely homework submission? These are confounding variables that I could never control for. Furthermore, there can be confounding variables that I do not know about.Please see my video below on how to conduct a Pearson r in SPSS and how to interpret the findings. JASP should be self-explanatory and the findings can be interpreted the same as SPSS.Remember, a Pearson r is a parametric test. I expect 2 tails. Their variables should both be numeric and paired.The assumptions of the Pearson r are:The 2 variables are measured are continuous variables (e.g. interval or ratio)The two variables are matched pairsThere is a linear relationship (we will inspect a scatter plot)There are no significant outliers (we will look at a box plot)There is bivariate (both variables) normality (KS test or Shapiro Wilks for smaller sample sizes). Our sample size is small, so I will use the Shapiro Wilk.Retrieved from: https://www.investopedia.com/terms/n/negative-correlation.aspLinks to an external site.The Direction of a Relationship The correlation measure tells us about the direction of the relationship between the two variables. The direction can be positive or negative.Positive correlation: In a positive relationship both variables tend to move in the same direction: If one variable increases, the other tends to also increase. If one decreases, the other tends to also. (Univsersitat de Valencia, n.d.)“Negative correlation, or inverse correlation, describes a situation where, with two variables, one variable increases in value while the other decreases” (Indeed Career Guide, 2023).See my video belowReferenceIndeed Career Guide. (2023, Feb). Negative Correlation: Definition and Examples (With Types). Retrieved from: https://www.indeed.com/career-advice/career-development/negative-correlation-definition-and-examplesLinks to an external site.Univsersitat de Valencia. (n.d.). Overview of Correlation. Retrieved from: https://www.uv.es/visualstats/vista-frames/help/lecturenotes/lecture11/overview-ovrh.htmlLinks to an external site.