Shopping class: Unit 0/Slide #1 Shopping for S-040: Introduction to Applied Data Analysis...

12
Shopping class: Unit 0/Slide #1 Shopping for S-040: Introduction to Applied Data Analysis freshspectrum.com © Eidelman, McIntyre, & Tutwiler, Harvard Graduate School of Education

Transcript of Shopping class: Unit 0/Slide #1 Shopping for S-040: Introduction to Applied Data Analysis...

Page 1: Shopping class: Unit 0/Slide #1 Shopping for S-040: Introduction to Applied Data Analysis freshspectrum.com © Eidelman, McIntyre, & Tutwiler, Harvard Graduate.

Shopping class: Unit 0/Slide #1

Shopping for S-040: Introduction to Applied Data Analysis

freshspectrum.com

© Eidelman, McIntyre, & Tutwiler, Harvard Graduate School of Education

Page 2: Shopping class: Unit 0/Slide #1 Shopping for S-040: Introduction to Applied Data Analysis freshspectrum.com © Eidelman, McIntyre, & Tutwiler, Harvard Graduate.

Shopping class: Unit 0/Slide #2© Eidelman, McIntyre, & Tutwiler, Harvard Graduate School of Education

A brief history of statistics at HGSE

S-012, full semester

OR

S-010Y, half-

semester

Fall semester: Spring semester:

S-030, full semester

… and then the following Fall:

S-052, full semester

Page 3: Shopping class: Unit 0/Slide #1 Shopping for S-040: Introduction to Applied Data Analysis freshspectrum.com © Eidelman, McIntyre, & Tutwiler, Harvard Graduate.

Shopping class: Unit 0/Slide #3© Eidelman, McIntyre, & Tutwiler, Harvard Graduate School of Education

S-040 is born!

Enter Katherine Masyn!

S-010Y + S-030 = S-040

Couldn’t we do this in one

semester, to give students

the opportunity to take an

intermediate methods class

in the Spring of

their first (or only) year?

Page 4: Shopping class: Unit 0/Slide #1 Shopping for S-040: Introduction to Applied Data Analysis freshspectrum.com © Eidelman, McIntyre, & Tutwiler, Harvard Graduate.

Shopping class: Unit 0/Slide #4© Judith D. Singer, Harvard Graduate School of Education

A simple survey

Thinking back over all the journal articles & research reports you read in your prior courses or professional work, in what percentage of papers did you carefully read the methods section?

All (100%)

Most (75-99%)

Many (50-74%)

Some (25-49%)

A Few (1-24%)

None (0%)

And across all the methods sections you read, in what percentage did you understand what the researchers did well enough to critically evaluate the credibility of the results?

All (100%)

Most (75-99%)

Many (50-74%)

Some (25-49%)

A Few (1-24%)

None (0%)

Do you plan on taking S-052, or another course which requires S-030 or S-040, in the spring?

Page 5: Shopping class: Unit 0/Slide #1 Shopping for S-040: Introduction to Applied Data Analysis freshspectrum.com © Eidelman, McIntyre, & Tutwiler, Harvard Graduate.

Shopping class: Unit 0/Slide #5

Are people less judgmental when they’re primed to feel clean?

Researchers found that people whoare primed to feel clean are lesssevere in their moral judgments.

But a replication study found noeffect of priming clean thoughtson severity of moral judgments.

?

the difference between not finding an associationand finding that there is no association

In class we’ll learn …how two different studies can reach different conclusions,even if both of them are well-designed

why we want you to wash your hands before fillingout your course evaluations

Page 6: Shopping class: Unit 0/Slide #1 Shopping for S-040: Introduction to Applied Data Analysis freshspectrum.com © Eidelman, McIntyre, & Tutwiler, Harvard Graduate.

Shopping class: Unit 0/Slide #6

Do all-nighters really not improve grades?

Time Magazine reports:“A new survey says those who never study all

night have slightly higher GPAs than those who do.”

Those who never stayed up all night

studying had an average GPA of 3.1

n=120 undergraduate students

Predictor: Had they stayed up all night studying, or had they not?

Outcome: Grade point average (possible range of 0-4)

Those who had stayed up all night

studying had an average GPA of 2.9

So, do all-nighters really not improve grades?

Maybe… BUT: Are we sure about the direction of the causal arrow?

© Singer, Eidelman, McIntyre, & Tutwiler, Harvard Graduate School of Education

Page 7: Shopping class: Unit 0/Slide #1 Shopping for S-040: Introduction to Applied Data Analysis freshspectrum.com © Eidelman, McIntyre, & Tutwiler, Harvard Graduate.

Shopping class: Unit 0/Slide #7

Was ABC News really beating NBC news in the ratings race?

9.69 million 9.65 million

So, was ABC really beating NBC?

Survey of 5,000

Maybe… BUT: These estimates are well within the limits of sampling error…

“… for the four-week period … Mr. Gibson’s broadcast was seen by an average of 9.69 million viewers a night, about 43,000 more

than the 9.65 million who watched Mr. Willams’ newscast.”

© Singer, Eidelman, McIntyre, & Tutwiler, Harvard Graduate School of Education

Page 8: Shopping class: Unit 0/Slide #1 Shopping for S-040: Introduction to Applied Data Analysis freshspectrum.com © Eidelman, McIntyre, & Tutwiler, Harvard Graduate.

Shopping class: Unit 0/Slide #8

What you’ll learn in S-040: The science and art of data analysis and linear regression

Unit 5:Inference for the Regression Model

Unit 6:Regression Assumptions:

Evaluating their Tenability

Unit 7:Transformations

to Achieve Linearity

Unit 8:The Basics of

Multiple Regression

Unit 9:Multiple Regression

in Depth: Special Topics

Unit 12:Interaction and

Quadratic Effects

Unit 10:Categorical Predictors I:

Dichotomies

Unit 11:Categorical Predictors II:

Polychotomies

Unit 14:Regression in Practice

and Common Extensions

Unit 4:Introduction to

Simple Linear Regression

Building a solid

foundation

Mastering the

subtleties

Adding additional predictors

Generalizing to other types of

predictors and effects

Pulling it all

together

Unit 1:The Basics: Categorical

Variables The basics

Unit 2:The Basics: Continuous

Variables

Unit 3:Correlation and

Causation

© Singer, Eidelman, McIntyre, & Tutwiler, Harvard Graduate School of Education

Page 9: Shopping class: Unit 0/Slide #1 Shopping for S-040: Introduction to Applied Data Analysis freshspectrum.com © Eidelman, McIntyre, & Tutwiler, Harvard Graduate.

Shopping class: Unit 0/Slide #9

By the end of this course, you will be able to

Quantify the association between two variables

Draw inferences from a sample to a wider population

Read, understand, and critique the methods sections of a wide variety of analyses

Do the same while holding constant, or controlling for, a third variable

Conduct a principled analysis of quantitative data from beginning to end

Communicate your results to a wide variety of audiences, including academic and lay

Take S-052 (Applied Data Analysis) or A-164 (Program Evaluation)

Test for and interpret statistical interactions

Page 10: Shopping class: Unit 0/Slide #1 Shopping for S-040: Introduction to Applied Data Analysis freshspectrum.com © Eidelman, McIntyre, & Tutwiler, Harvard Graduate.

Shopping class: Unit 0/Slide #10© Judith D. Singer, Harvard Graduate School of Education

How you’ll spend your time in S-040, Part I:What we’ll do in class

I. Research Questions and Data Sets

• What is the relationship between median home values and MCAS pass rates?

• Did Al Gore really lose the 2000 Presidential Race in Florida?

• Are there racial disparities in the lengths of criminal sentences among convicted felons?

• … and many more

Lectures with your questions:Active participation is encouraged (and often a great part of the learning experience)

II. Delve into the new statistical content that the RQs (and the unit)

demands• What aspect of the model do we need to

learn more about?• How do we represent this aspect of the

model algebraically & graphically?• What assumptions are we making (and

how do we evaluate whether these make sense?)III. Interpreting & presenting results

• How do we interpret computer output?• What conclusions can we draw—and

what conclusions don’t necessarily follow?

• How do we write up our results—in words, graphs, tables, PowerPoints?

• How do we communicate results to both technical and non-technical audiences?

Each unit has a three-part structure

Note-taking:On laptop or printouts of

handouts

Please be courteous:

No cellphones; keep email, web surfing,

Facebook, etc., to an absolute minimum

(so as not to distract classmates)

Page 11: Shopping class: Unit 0/Slide #1 Shopping for S-040: Introduction to Applied Data Analysis freshspectrum.com © Eidelman, McIntyre, & Tutwiler, Harvard Graduate.

Shopping class: Unit 0/Slide #11

How you’ll spend your time in S-040, Part II:What you’ll do outside of class

Assignments

• Six carefully constructed homework assignments, each consisting of a RQ, data set, & questions that guide you through a complete analysis (~50% grade)

• Several problem sets, which will be relatively short in comparison to the homework assignments (~25% grade)

• One final project that gives you a chance to pull together all your component skills into a polished professional product (~25% grade)

Partnerships, Individuals, and Group Work

• Work in study groups as you’d like, but write and submit HWs and the final with your one partner only

• We will let you know which problem sets are individuals-only and which ones require a partner

TFs will hold 1.5-hour labs each

week, to support Stata coding and core concepts of

the course

Course website: http://isites.harvard.edu/icb/icb.do?keyword=k106039

Very little required reading:

… available in your iPac …

© Singer, Eidelman, McIntyre, & Tutwiler, Harvard Graduate School of Education

Statistical computing

with TF support

Page 12: Shopping class: Unit 0/Slide #1 Shopping for S-040: Introduction to Applied Data Analysis freshspectrum.com © Eidelman, McIntyre, & Tutwiler, Harvard Graduate.

Shopping class: Unit 0/Slide #12

Eight things you should do before the first class meeting, next Thursday

1. Make sure you email Joe to get

instructor permission to take this course. If you

are a cross-registrant, check in

with us on next steps.

7. Decide how you want to access

Stata• Read through the

“Statistical computing” section of the syllabus.

• Think about whether it makes sense for you to purchase a license.5. Familiarize yourself

with the S-040 iSite• Bookmark the site.

• Read the syllabus — it includes many more details and is our learning contract.

8. Bring the first handout to class

(print or download)• We’ll be posting the 1st

handout to the website by next Wednesday morning.

• You don’t have to read it; just be sure to bring it.

3. Register for the course

4. Read the School’s policy on plagiarismAll written work submitted is to be in your own words

6. Read “Best Practices in S-

040”Helpful advice from former

students and TFs

2. Complete and submit the sign up sheet (if by email,

send to Hadas)

© Singer, Eidelman, McIntyre, & Tutwiler, Harvard Graduate School of Education