More detailed notes about what is required in each section of the report.
Some parts are relevant to those doing the Experimental design (IV = Routes of Persuasion/ Central & Peripheral) and some parts more relevant to those doing the Quantitative Observational Design (IV = Extent of prior tobacco exposure)
Those doing Quantiotative Observational should read the section on Correlation in the PDF below. In your results section you will need to ctalk about the meaning of correlation and why you have used it. The notes give you the necessary guidelines to do that.
Excel will do the linear regression required to fit the line to the graph(s) of Exposure Levels (IV on the x axis) versus the dependent variables (Attitude Scores in Self and to Others). If you sre struggling with excel produce a paper graph for the time being so that you can see twhat the graph looks like in order to properly discuss the relationship between exposure level to tobacco and effects on attitudes (is there a correlation?).

Correlation and Linear Regression - What is it about? Correlation is the statistical measurement of the relationship [link] between two variables. So, for example: if you were to ask the question “is there a predictable change in the attitudes of Self or towards Others concerning smoking as a function of the extent of prior experience/exposure to tobacco smoking”; then, you are really asking, are the two correlated – is the dependent variable co-related to the independent variable? So perhaps the unusual thing for you to consider here is the fact that both measures are changing. You are more familiar with making fixed comparisons; eg, for a given measure you compare a (single) mean for a group to the (single) mean of another group. The nature of relationships between variables may be complex- but the simplest type is a linear relationship – as one variable changes there is a proportionate change in the other. When you plot one versus the other, a straight line results. The “straightness” of a line, or the fit of a line to data, is called linear regression – and Excel does it for you on the press of a button! 1. CONSTRUCT A SCATTER PLOT – select the appropriate data array in excel and then select scatter plot 2. EDIT THE EXCEL GRAPHING OPTIONS FROM THE DROP DOWN choices– choose no lines, then right click data series, add trendline, add R2 .... voila, EXCEL has done all of the work for you. You now need to explain what it means. The statistical measure is called the correlation coefficient and is represented by the letter R2. As said, Excel calculates if for you. Correlation is a statistical measurement of the relationship between two variables. Possible correlations range from +1 to –1. A zero correlation indicates that there is no relationship between the variables. A correlation of –1 indicates a perfect negative correlation, meaning that as one variable goes up, the other goes down. A correlation of +1 indicates a perfect positive correlation, meaning that both variables move in the same direction together. Just a simple rule of thumb for the present, if a R2 is between -0.6 and +0.6 then there is probably no statistically significant correlation (or the data is poor, not enough data, outliers, etc). By the way, an R2 of greater than 0.8 or less than -0.8 is a very significant correlation (1, or -1 is a perfect correlation).

## RESEARCH PROJECT

Notes to assist in writing the Research ReportPlease note this page has not been updated as yet for the 2011 projectUPDATED DATA FILE- includes linear regression examplesSome parts are relevant to those doing the Experimental design (IV = Routes of Persuasion/ Central & Peripheral) and some parts more relevant to those doing the Quantitative Observational Design (IV = Extent of prior tobacco exposure)

Those doing Quantiotative Observational should read the section on Correlation in the PDF below. In your results section you will need to ctalk about the meaning of correlation and why you have used it. The notes give you the necessary guidelines to do that.

Excel will do the linear regression required to fit the line to the graph(s) of Exposure Levels (IV on the x axis) versus the dependent variables (Attitude Scores in Self and to Others). If you sre struggling with excel produce a paper graph for the time being so that you can see twhat the graph looks like in order to properly discuss the relationship between exposure level to tobacco and effects on attitudes (is there a correlation?).

Correlation and Linear Regression - What is it about?

Correlation is the

statistical measurementof the relationship [link] between two variables. So, for example: if you were to ask the question “is there a predictable change in the attitudes of Self or towards Others concerning smoking as a function of the extent of prior experience/exposure to tobacco smoking”; then, you are really asking, are the two correlated – is the dependent variable co-related to the independent variable?So perhaps the unusual

thingfor you to consider here is the fact that both measures are changing. You are more familiar with making fixed comparisons; eg, for a given measure you compare a (single) mean for a group to the (single) mean of another group.The nature of relationships between variables may be complex- but the simplest type is a linear relationship – as one variable changes there is a proportionate change in the other. When you plot one versus the other, a straight line results. The “straightness” of a line, or the fit of a line to data, is called linear regression – and Excel does it for you on the press of a button!

1. CONSTRUCT A SCATTER PLOT – select the appropriate data array in excel and then select

scatter plot2. EDIT THE EXCEL GRAPHING OPTIONS FROM THE DROP DOWN choices– choose no lines, then right click data series, add trendline, add R2 .... voila, EXCEL has done all of the work for you. You now need to explain what it means.

The statistical measure is called the

correlation coefficientand is represented by the letter R2. As said, Excel calculates if for you.Correlation is a statistical measurement of the relationship between two variables. Possible correlations range from +1 to –1. A zero correlation indicates that there is no relationship between the variables. A correlation of –1 indicates a perfect negative correlation, meaning that as one variable goes up, the other goes down. A correlation of +1 indicates a perfect positive correlation, meaning that both variables move in the same direction together.

Just a simple rule of thumb for the present, if a R2 is between -0.6 and +0.6 then there is probably no statistically significant correlation (or the data is poor, not enough data, outliers, etc). By the way, an R2 of greater than 0.8 or less than -0.8 is a very significant correlation (1, or -1 is a perfect correlation).

Summary 2009 2010 Persuasion Project Data Incl Linear regression.xlsx

Not for the Research Report but a Formative activity to do during the holidays.

Visit your local video store for: The Devil Wears Prada

This is a worksheet to complete while/after watching the movie.