Assignment on How to evaluate inferential and descriptive statistics

Module #6:
How to evaluate inferential and descriptive statistics
Must be turned in before: June 26, 11:59 PM
a. To review what descriptive and inferential statistics are, why they are important to
learn, and examples of how they are used:
1. Watch Lynda.com’s (2010) video, “Understanding Descriptive and
Inferential Statistics.”
2. Read Laerd Statistics’ (no date) article, “Descriptive and Inferential
Statistics.”
3. Read a section of Wikipedia’s (2017) entry, “Descriptive Statistics.”
4. Read Statistics HowTo’s (2014) article, “Inferential Statistics: Definition,
Uses.”
b. To really make sure you understand the difference between descriptive and
inferential statistics and what each is used for, watch StatsLectures (2010) video,
“The Basics: Descriptive and Inferential Statistics.”
1. At this point, you should be clear on the difference between descriptive
and inferential statistics and the common uses for both types of statistics.
2. If you’re not clear, you might want to re-read the above articles and rewatch the videos.
3. You might also want to review how to write a Five-Paragraph ExamplesStyle Essay, by re-watching the latter part of “The Five-Paragraph
Model”
WRITING ASSIGNMENT
c. Write one five-paragraph essay where you use three examples to support the
thesis “Descriptive statistics are useful.” Remember that descriptive statistics can
be graphs and figures, as well as means and modes.
a hook and a Thesis Statement.
2. Check your Thesis Statement to make sure that it summarizes your
three examples.
3. Check your essay to make sure it has three Supporting Paragraphs.
4. Check each of your three Supporting Paragraphs to make sure each one
has a Topic Sentence, three or so Supporting Sentences, and
a Conclusion Sentence.
5. Check your essay to make sure it has a Conclusion Paragraph.
Part 1: Inferential and descriptive statistics
6. Check your Conclusion Paragraph to make sure it has a sentence that
d. In the same document as your first essay, write another five-paragraph essay
(again, using examples), this time to support the thesis “Inferential statistics are
useful.”
a hook and a Thesis Statement.
2. Check your Thesis Statement to make sure that it summarizes your
three examples.
3. Check your essay to make sure it has three Supporting Paragraphs.
4. Check each of your three Supporting Paragraphs to make sure each one
has a Topic Sentence, three or so Supporting Sentences, and
a Conclusion Sentence.
5. Check your essay to make sure it has a Conclusion Paragraph.
6. Check your Conclusion Paragraph to make sure it has a sentence that
7. Save the entire document as a PDF (which should contain two fiveparagraph essays) and name the
file YourLastname_DescriptiveInferentialEssays.pdf.
How to evaluate inferential and differential statistics”
e. If you ever wonder why you will repeatedly practice skills, such as writing fiveparagraph essays, in different contexts throughout this course, consider the
words of William James, who is widely considered the father of modern
psychology!
a. To become familiar with some of the ways that descriptive and inferential
statistics can be used to deceive people, read Chapters 2 through 6 of (a
slender!) book titled How to Lie with Statistics by Darrell Huff.
1. NOTE: This book was published in 1954; therefore, the examples are from
the 1940s and early 1950s. However, it’s still a beloved book (e.g., it’s
recommended reading in a college physics class), despite its age.
2. Chapter 2 explains the deception caused by indiscriminately referring to
the mean, median, and mode (i.e., three central-tendency descriptive
statistics) as “the average.”
3. Chapter 3 explains the deception caused by random variation and the
solutions provided by inferential statistics.
4. Chapter 4 explains the deception caused by differences that aren’t
meaningful.
Part 2: How to lie with statistics
5. Chapters 5 and 6 explain deception by graphs and figures.
b. When reading these chapters, jot down your three favorite deceptions. For
example, you might choose as one of your favorite deceptions the hypothetical
real estate agent’s deceptive use of a neighborhood’s “average” income in
Chapter 2.
WRITING ASSIGNMENT
c. Create a teaching document to teach your three favorite deceptions to other
people.
1. You need to choose an audience for your teaching document. Your
choices are (1) other college students; (2) middle-school students (age 12
to 14); or (3) older adults (over age 60).
2. You need to choose a medium for your teaching document. Your choices
are (1) a powerpoint; (2) an infographic (what’s that?); or (3) a comic strip
(e.g., The Nib’s).
3. You need to save your teaching document as a PDF,
named YourLastname_StatsDeception.pdf.
d. Save your teaching document to the discussion board forum “Assignment #6,
Part 2: Deception Teaching Document.”
a. To learn (or be reminded) what effect size is and why it’s important to report it in
scientific articles:
1. Read Sullivan and Feinn’s (2012) article, “Using Effect Size—or Why the
P Value Is Not Enough.”
2. Sullivan and Feinn’s (2012) article might be harder to read than other
articles you’ve read in this course. But try to understand it at least at a
superficial level. Feel free to Google terms that you don’t know.
WRITING ASSIGNMENT
b. Now it’s time to apply Sullivan and Feinn’s article to an actual scholarly scientific
article.
1. Choose three of the scholarly articles you cited in Assignment #4 (from
either part 1 or part 2). Pay particular attention to the abstract and results
sections.
2. Evaluate the three articles as follows:
▪ Read the abstract and determine what the most important finding
was (the primary hypothesis, which is usually spelled out in the title
of the article).
Part 3: Effects sizes
▪ Go to the results section and find the paragraph that describes the
statistical analysis of the results for the primary hypothesis. Take a
screen shot of that paragraph.
▪ Try to find the statistic that indicates the effect size. Possible
indicators of effect size are Cohen’s d, the 95% confidence interval,
r squared or eta squared.
▪ Create a word document or PDF reporting the following information
for each study:
a. The title of the study, with a link to the study hyperlinked in
the title
b. What the study was testing (the research question or
hypothesis
c. A screenshot of the pertinent results section
d. The statistical test used to test the hypothesis, the value of
that statistic + the p value.
e. The statistic used to report the effect size and its value.
Example:
Study 1
a. Title: Threat-induced modulation of hippocampal and striatal memory systems
during navigation of a virtual environment
b. This study tested the influence of anticipatory anxiety (the threat of receiving
c.
d. A paired-sample t-test was used to test the hypothesis; t = 2.688; p = .0118.
e. Cohen’s d was used to report the effect size; d = .491.
Attach your .pdf or .doc to the discussion board forum “Assignment #6, Part 3: Effect
Sizes.”
polls and surveys, read Langers’ (2008) article, “ABC News’ Guide to Polls &
Public Opinion.”
1. Make sure you understand the article’s answer to the concern that “[the
pollsters] never call me.”
2. Make sure you understand the article’s answer to the concern that
“nobody I know says that.”
b. To better understand “margin of error” means in statistics:
1. Read Rumsey’s (no date) article, “How to Interpret the Margin of Error
in Statistics.”
2. Make sure you understand the difference between sampling a population
and surveying (or polling) an entire population.
3. Make sure you understand what a margin of error means in a public
opinion poll or survey.
c. To better understand the relation between sample size and margin of error:
1. Read Hunter’s (no date) article, “Margin of Error and Confidence Levels
2. Make sure you understand what it means to calculate a margin of error at
a 95% confidence level.
3. Make sure you understand the relation between sample size and margin
of error.
d. Choose three of the following topics for which Gallup has recently conducted a
public opinion poll. Then, within each of the three topics you’ve chosen,
read one of the listed reports.
1. Employment
▪ Harter and Adkins (2015): “Engaged Employees Less Likely to
Have Health Problems“
▪ Newport (2017a) “Email Outside of Working Hours Not a Burden
to US Workers“
▪ Newport and Dugan (2017): “Americans Still See Manufacturing
as Key to Job Creation.”
▪ Newport (2018a): “Average American Predicts Retirement Age
of 66“
Part 4: Statistics and Public Media
▪ Swift (2017a): “Most U.S. Employed Adults Plan to Work Past
Retirement Age“
2. Finance
▪ Newport (2017b): “Young, Old in US Plan on Relying More on
Social Security“
▪ Jones (2017a): “Worry About Hunger, Homelessness Up for
Lower-Income in US“
▪ Norman (2017): “Financially Stressed in US Now Prefer Saving
to Spending“
▪ Jones (2017b): “Half of Non-Homeowners Expect to Buy Homes
in Five Years“
▪ Newport (2018b): “Americans’ Views of Their Spending and
Saving“
3. Well-Being
▪ Rigoni and Nelson (2016): “Millennials Want Jobs That Promote
Their Well-Being“
▪ Witters (2017a): “Hawaii Leads US States in Well-Being for
Record Sixth Time“
▪ Witters (2017b): “Naples, Florida, Remains Top US Metro for
Well-Being“
4. National Policy
▪ McCarthy (2017a): “US Support for Gay Marriage Edges to New
High“
▪ McCarthy (2017b): “Americans More Positive about Effects of
Immigration“
▪ Swift (2017b): “More Americans Say Immigrants Help Rather
Than Hurt Economy“
▪ Reinhart and Ray (2018): “Record Unhappiness with Women’s
Position in U.S.“
5. Higher Education
▪ Auter (2018): “Half of College Students Say Their Major Leads
to a Good Job“
▪ Maturo (2017): “One in Three Veterans Consult Coworkers