EWA Instructions
Papers must be typed (with the exception of the prediction equation if you do not know how to use equation editor).
Paper must be written independently (if your paper sounds a lot like another student’s paper or has the exact
same results as another student’s paper you will both receive a zero and you will be reported to academic integrity).
Papers must be printed out and turned in in person to me. Do not email me your paper.
Papers can be single spaced and twosided (or any other format is okay as well).
How to write your paper?



You want to write this paper so that someone without statistical knowledge understands what you did and
what the results show.
Make sure you mention how each of the objects were identified and
Follow the EWA Tips for Lab 4 carefully.
Biggest issue with EWA is students not including everything outlined on the tip sheet.
You may use headers and work through the tips one at a time. (Treat this as a checklist but make sure that
your paper is written as a paper not as bullet points. In other words, the information under the headers
should be in paragraph form unless it is a table or a graph)
Don’t forget anything outlined on the Tips sheet as this is what is used for grading.
o For example, under results it says that a roadmap is required. All you have to do is make a paragraph
with one sentence per table you are going to include in this section. (all sentences go before the tables
in a paragraph). This is not hard to do but if you don’t include it you get points off.
Don’t forget the appendix – this can be handwritten.
STAT 201: Tips on EWA for Lab Session 4
Style: The text of the EWA must be typed. Some neat, handwritten labels and/or numbers on figures are permitted.
Independent Writing: The EWA is done independently, not in your lab group. Each student will use their own individual results so
there are no two papers that have the same analysis results.
Overview of Paper Sections:
Title: Make sure that your EWA has a descriptive title – EWA or Lab 4 or ‘You Seem so Distant’, the book’s title is unacceptable.
Abstract: This is a stand alone section. It is written to give the reader a brief synopsis of the experiment and some general findings.
This is written for a person who is doing research on a topic possibly related to your experiment. Be sure to be specific, yet brief.
Even though the abstract comes first in your paper, it is written last. It is a summary of your paper, and is difficult to write well
without having written your paper. There will be some redundancy from other parts of the paper.
In the abstract be sure to include:
→ The purpose of the experiment
→ A brief description of the data collection and analysis (exclude fine details which will be covered in your paper)
→ Your conclusion – Can we effectively use guessed distances to predict true distances ?
Materials and Methods: This section should include:
→ Description of experiment’s purpose (Hint: what were you trying to decide?)
→ State the materials that were used.
→ Description of basic experiment – describe how the experiment was setup
→ Description of statistical methods. Elaborate on each method – do not just list them! Describe not only what methods were
used but why they were used in the context of this regression analysis. Do NOT give any numbers, just describe the analysis
techniques that were performed. Someone who knows nothing of the experiment should be able to repeat the entire analysis based
on your description. Methods used include: Scatterplot, line y = x to determine type of guesser, correlation coefficient, simple
linear regression (including line of bit fit and prediction equation creation), and residuals.
Results: In this section, Tables and Figures are presented without interpretation. There should be a brief "road map" paragraph
explaining to the reader what Tables and Figures are found in this section.
Results section should include:
→ Note that a roadmap PARAGRAPH (not a list) is required to describe the tables and figures in the results section. This should
appear at the beginning of the Results section. The roadmap paragraph consists of approximately one sentence per table
explaining to the reader what information is contained in the following tables. Do NOT tell why the table is important – you have
already covered this in the materials and methods section.
→ Include a scatterplot of the data (do not include the 11 th item here) and include on the graph a superimposed line for y = x.
→ Include all output generated by StatCrunch from the simple linear regression analysis (note this will be the “Simple linear
regression results” and the scatterplot with the fitted line on it).
→ Include a table that contains information for the 11th landmark: prediction equation (this can be handwritten if you can’t figure out
how to write it in word), your guessed distance to the 11 th landmark; the predicted distance to the 11th landmark using the
prediction equation, the true distance to the 11th landmark, and the residual calculated for the 11th landmark.
Discussion: In this section, make sure that you do the following:
→ Address the purpose of the experiment (again).
→ Clearly identify the independent and dependent variables in this analysis.
→ Discuss what conclusions can be reached from the scatterplot with the line y = x on it. This should include what type of
relationship there is in the data, if there are any outliers, and what type of guesser you are. Provide justifications/explanations for
each.
→ Discuss what conclusions can be reached from the correlation coefficient, clearly state what happens to y as x increases (don’t use
“x” and “y” in the sentence), discuss whether any possible correlation implies a cause and effect relationship between the two
varaibles.
→ Discuss the results of the simple linear regression analysis and how the prediction equation was created:
o What does the line of best fit on the scatterplot show and how close do the dots fall to the line? Elaborate a little on
this.
o Give and interpret the slope of the line of best fit.
o Give and interepret the yintercept of the line of best fit. Comment on whether this interpretation was
meaningful/practical (if it isn’t then interpret it anyway)
o Give the prediction equation and comment on how predictions are made using this model (make sure to mention the
relevant range).
→ Discuss the difference between the predicted true distance of the 11 th landmark and the guessed distance.
→ Discuss the residual of the 11th landmark and interpret the value you calculated for it.
→ Comment on if you think the model was good and interprete rsquared.
→ Discuss any shortcomings of the experiment – should there have been additional steps in the protocol of the experiment to
eliminate bias, etc.
→ Ideas for further experimentation: what could be done to change the experiment? Do not include ideas that are totally unrelated to
a regression analysis!
Appendix: This section can be handwritten. This section must contain the following:
→ Table that contains the guessed distances and true distances.
→ Calculation of the predicted distance to the 11th landmark. You must show the formula used and the end result.
→ Calculation of the residual of the 11th landmark. You must show the formula used and the end result.
Rubric for STAT 201 EWA – Spring 2019 Edwards
Title – 1
Entire Report – 3
Abstract – 2.5
0.5Purpose of experiment
3Spelling and grammar
1Brief description of data collection and analysis
1Conclusion and supporting results
Materials & Methods – 9
0.5 Descripition of experiments purpose
0.5 – Materials used
1 – Set up of the lab.
3Description of sample statistics
1
Information about relationship based on scatterplot
1
Determining type of guesser
1
Information about relationship based on correlation coefficient.
4Explanation of Statistical methods
1
Determining Line of Best Fit.
1
Explanation of how prediction equation is created.
0.5 Use of prediction equation.
0.5
relevant range
1
residuals
Results – 5
1Roadmap referring to tables and figures
4Data Presentation
1
Scatterplot with y = x line
1
Simple linear regression results from StatCrunch – prediction equation, r, rsquared, slope and intercept values.
1
Scatterplot with line of best fit
1
Table with prediction equation, guessed 11 th, predicted 11th, true 11th, residual.
Discussion – 6.5
0.25 Purpose
0.5 Independent vs Dependent variables.
0.5 Relationship based on scatterplot with justification
0.25 Outliers
0.5 type of guesser with justification.
0.5 Relationship based on correlation coefficient with justification.
0.25 Tendency of y as x increases
0.25 Does correlation imply causation?
0.5 Line of best fit – where do the dots fall in relation to line.
0.5 Interpretation of slope.
0.5 Interpreation of yintercept and was it useful/meaningful.
0.5 Explaining how prediction equation is used and mention relevant range
0.25 difference between guessed distance and predicted distance for 11 th landmark.
0.25 Interpret the residual calculated for the 11th landmark.
0.5 – Goodness of fit of the model / interpret rsquared.
0.25Any shortcomings of the experiment?
0.25Ideas for future experimentation
Appendix – 3
1
Table of data Collected
1
Calculation of predicted distance for the 11 th landmark
1
Residual calculations for object 11