powerful but invisible: social psychological processes ... · coach wooden and me. this need to be...

Powerful but invisible:Social psychological processes shaping student success

Geoffrey Cohen, Stanford University

Presentation prepared for Boston University5th Annual Advising SymposiumCivic Engagement: Advising in a Pluralistic SocietyFebruary 2, 2018

The same situation can be perceived differently.

I decided it would be better not to say I was Jewish, even though therewas no obvious Anti-Semitism at my school. The Jewish boys, even thepopular ones, just seemed to have a subtly charged field around them,an air of apartness that did not emanate from any quality of their own,but from the school—as if a guardian spirit had risenfrom the walkways and weathered stone and breathed that apartnessupon them.

Paraphrased from T. Wolff, Old School

Feeling like an outsider:A common predicament

Transfer studentFirst generation college studentLGBQ studentConservative student on liberal campusAthlete


Transfer studentFirst generation college studentLGBQ studentConservative student on liberal campusAthlete in class


Transfer studentFirst generation college studentLGBQ studentConservative student on liberal campusAthlete in classAcademic in sports

The same classroom can be experienced more differently than different classrooms.

Hannah Sung Cosmos, University of Pittsburgh

Tanner LeBaron Wallace, University of Pittsburgh

The same classroom can be experienced more differently than different classrooms.

0

0.2

0.4

0.6

0.8

1

1.2

Between classroom variance Within classroom variance

Students' perceptionsStudents' learning

Psychology matters for motivation, performance, and health.

Some essence drained out of my body.”-Kareem Abdul-Jabbar, Coach Wooden and Me

This need to be always on guard was an unmeasured expenditure of energy..

-T. Coates, Between the World and Me

The function, the very serious function of racism, is distraction. It keeps you from doing your work. It keeps you explaining, over and over, your reason for being.

-Toni Morrison, Toni Morrison, Black Studies Center Dialogue

“I knew I was just as intelligent as everyone else . . . . For some reason I didn’t score well on tests. Maybe I was just nervous. There’s a lot of pressure on you, knowing that if you fail, you fail your race.”

-Rodney Ellis, African American State Senator (Texas), 1997

Some essence drained out of my body.”-Kareem Abdul-Jabbar, Coach Wooden and Me

This need to be always on guard was an unmeasured expenditure of energy.

-T. Coates, Between the World and Me

The function, the very serious function of racism, is distraction. It keeps you from doing your work. It keeps you explaining, over and over, your reason for being.

-Toni Morrison, Toni Morrison, Black Studies Center Dialogue

“I knew I was just as intelligent as everyone else . . . . For some reason I didn’t score well on tests. Maybe I was just nervous. There’s a lot of pressure on you, knowing that if you fail, you fail your race.”

-Rodney Ellis, African American State Senator (Texas), 1997

Stereotype threat experiment:When White men can’t jump

0

0.2

0.4

0.6

0.8

1

1.2

1.4

White Experimenter Black Experimenter

Impr

ovem

ent f

rom

Jum

p 1

to J

ump

2 (in

ches

)

Stereotype threat experiment:When White men can’t jump

0

0.2

0.4

0.6

0.8

1

1.2

1.4

White Experimenter Black Experimenter

Impr

ovem

ent f

rom

Jum

p 1

to J

ump

2 (in

ches

)

*

* P < 0.05

Test Performance(Number Correct)

2

4

6

8

10

12

"Test is measure ofability"

"Test unrelated toability"

African AmericansWhites

Math Test Performance Among Men and Women(Percent Correct)

0%

10%

20%

30%

40%

50%

Neutral Commercial Stereotypical Commercial

MenWomen

Source: Davies, Spencer, Quinn, Gerhardstein, PSPB

- +

-

Social Reality

Vigilance

Cues confirm threat Cues disconfirm threat

Motivation, performance

+

- +

Social Reality

Vigilance



- +

-

Social Reality

Vigilance



+

Changing the social environment

• Eliminating bias

• Fostering and valuing diversity

Changing the social environment

• Eliminating bias

• Fostering and valuing diversity

May require innovations to “decision architecture.”

Color Consciousness + Accountability Intervention:

Discrimination BEFORE Intervention

0

-5

-10

Military Promotion Rates: BEFORE intervention:Minority Males Relative to White Males

Perc

ent D

iffer

ence

Capt. Major Lt. Col. Col.

NS

-4.2-2.2

-5.9

Source: Career Progression of Minority and Women Officers, 1999

0

-5

-10

Army

Minority Males Relative to White Males

Percent Difference


0

-5

-10

Army

Percent Difference


0

-5

-10

Military Promotion Rates: BEFORE intervention:


Percent Difference


NS

-4.2

-2.2

-5.9

Source: Career Progression of Minority and Women Officers, 1999

0

-5

-10

Military Promotion Rates AFTER intervention:Minority Males Relative to White Males

Perc

ent D

iffer

ence



Discrimination AFTER Intervention

0

-5

-10

Army


Percent Difference


0

-5

-10

Military Promotion Rates AFTER intervention:


Percent Difference


0

-5

-10

Military Promotion Rates AFTER intervention ENDED:Minority Males Relative to White Males

Perc

ent D

iffer

ence



Discrimination AFTER Intervention ENDED

0

-5

-10

Military Promotion Rates AFTER intervention ENDED:


Percent Difference


Counteracting the costs of vigilance


• The power of relationships


• The power of relationships

• The power of the encounter

My family had no money when I was young. We were always on theverge of loosing the farm.

One day, my elementary school teacher, Mrs. Zinn, kept me after class.She was big. She was commanding voice. She scared people.

She said I could do more than become a farmer. She saw I had abilities Idid not know I had. She gave me extra work.

She was the spark who made me believe in me.

Mrs. Zinn was a counter-weight to the harsh messages I got aboutmyself elsewhere. A person that has their self-worth pounded on dayafter day needs proof that there is something of value still in them.

I had editors—more teachers—and these were thefirst white people I’d ever really known on anypersonal level. They defied my presumptions.They were afraid neither for me nor of me.Instead they saw in my unruly curiosity andsoftness something that was to be treasuredand harnessed.

Paraphrased from T. Coates, Between the World and Me

Catalysts for change

• Trust

• Belonging

• Growth mindset

• Self affirmation

Creating turning points:When social influence lasts

Targeted, tailored, and timely influence

Right person, right message, right time

Cohen, Garcia, & Goyer (2017)

Building trust

Perceived Instructor Bias

1

2

3

4

5

UnbufferedCriticism

Criticism +Positive Buffer

Criticism + High Standards

+ Assurance

African American StudentsWhite Students

Scale range: 1-7

2 pages of critical feedback

Chart2

Unbuffered CriticismUnbuffered Criticism0.350.350.260.26

Criticism + Positive BufferCriticism + Positive Buffer0.350.350.270.27

Criticism + High Standards + AssuranceCriticism + High Standards + Assurance0.280.280.280.28

African American Students

White Students

3.47

2.31

2.73

2.01

2.04

2.47

Fig 1 data

WeeksBluegillPumpkinseedGreen Sunfish

181417

31979

51458

76318

910619

111247

13532

15433

Column A = x-axis values

Fig 1 (scatterplot)

111

333

555

777

999

111111

131313

151515

Bluegill

Pumpkinseed

Green Sunfish

Time (weeks)

Dry Weight of Food (mg)

Figure 1. Weekly changes in dry weight of food eaten by three sunfish species in a pond experiment.

8

14

17

19

7

9

14

5

8

6

3

18

10

6

19

12

4

7

5

3

2

4

3

3

Fig 2 data

YearPopulation Size

198010

198115

198225

198320

198440

1985100

1986180

1987210

1988270

1989285

1990295


Fig 2 (line graph)

1980

1981

1982

1983

1984

1985

1986

1987

1988

1989

1990

Population Size

Year

N (Population Size)

Figure 2. Population growth of mule deer in Glacier County.

10

15

25

20

40

100

180

210

270

285

295

Fig 3 data

Unbuffered CriticismCriticism + Positive BufferCriticism + High Standards + Assuranceblack se'swhites se's

African American Students3.472.732.040.350.26

White Students2.312.012.470.350.27

0.280.28

Column A = x-axis labels

Fig 3 data

0.350.350.260.26

0.350.350.270.27

0.280.280.280.28


White Students

Fig 3 (bar graph)

African American StudentsAfrican American Students2.040.35

White StudentsWhite Students2.470.35

Unbuffered Criticism

Criticism + Positive Buffer

Nutrient

Percent Increase Compared to Control

Figure 3. Responses of lake algae to addition of nutrients. Error bars are 95% confidence limits.

N = 50 lakes.

3.47

2.73

2.31

2.01


1

2

3

4

5

UnbufferedCriticism



+ Assurance


Scale range: 1-7

Chart2





White Students

3.47

2.31

2.73

2.01

2.04

2.47

Fig 1 data


181417

31979

51458

76318

910619

111247

13532

15433


Fig 1 (scatterplot)

111

333

555

777

999

111111

131313

151515

Bluegill

Pumpkinseed

Green Sunfish

Time (weeks)



8

14

17

19

7

9

14

5

8

6

3

18

10

6

19

12

4

7

5

3

2

4

3

3

Fig 2 data

YearPopulation Size

198010

198115

198225

198320

198440

1985100

1986180

1987210

1988270

1989285

1990295


Fig 2 (line graph)

1980

1981

1982

1983

1984

1985

1986

1987

1988

1989

1990

Population Size

Year

N (Population Size)


10

15

25

20

40

100

180

210

270

285

295

Fig 3 data



White Students2.312.012.470.350.27

0.280.28


Fig 3 data

0.350.350.260.26

0.350.350.270.27

0.280.280.280.28


White Students

Fig 3 (bar graph)





Nutrient



N = 50 lakes.

3.47

2.73

2.31

2.01


1

2

3

4

5

UnbufferedCriticism



+ Assurance


Scale range: 1-7

“Overall, nice job. Your enthusiasm really shows through…” +2 pages of critical feedback

Chart2





White Students

3.47

2.31

2.73

2.01

2.04

2.47

Fig 1 data


181417

31979

51458

76318

910619

111247

13532

15433


Fig 1 (scatterplot)

111

333

555

777

999

111111

131313

151515

Bluegill

Pumpkinseed

Green Sunfish

Time (weeks)



8

14

17

19

7

9

14

5

8

6

3

18

10

6

19

12

4

7

5

3

2

4

3

3

Fig 2 data

YearPopulation Size

198010

198115

198225

198320

198440

1985100

1986180

1987210

1988270

1989285

1990295


Fig 2 (line graph)

1980

1981

1982

1983

1984

1985

1986

1987

1988

1989

1990

Population Size

Year

N (Population Size)


10

15

25

20

40

100

180

210

270

285

295

Fig 3 data



White Students2.312.012.470.350.27

0.280.28


Fig 3 data

0.350.350.260.26

0.350.350.270.27

0.280.280.280.28


White Students

Fig 3 (bar graph)





Nutrient



N = 50 lakes.

3.47

2.73

2.31

2.01


1

2

3

4

5

UnbufferedCriticism



+ Assurance


Scale range: 1-7

Chart2





White Students

3.47

2.31

2.73

2.01

2.04

2.47

Fig 1 data


181417

31979

51458

76318

910619

111247

13532

15433


Fig 1 (scatterplot)

111

333

555

777

999

111111

131313

151515

Bluegill

Pumpkinseed

Green Sunfish

Time (weeks)



8

14

17

19

7

9

14

5

8

6

3

18

10

6

19

12

4

7

5

3

2

4

3

3

Fig 2 data

YearPopulation Size

198010

198115

198225

198320

198440

1985100

1986180

1987210

1988270

1989285

1990295


Fig 2 (line graph)

1980

1981

1982

1983

1984

1985

1986

1987

1988

1989

1990

Population Size

Year

N (Population Size)


10

15

25

20

40

100

180

210

270

285

295

Fig 3 data



White Students2.312.012.470.350.27

0.280.28


Fig 3 data

0.350.350.260.26

0.350.350.270.27

0.280.280.280.28


White Students

Fig 3 (bar graph)





Nutrient



N = 50 lakes.

3.47

2.73

2.31

2.01


1

2

3

4

5

UnbufferedCriticism



+ Assurance


Scale range: 1-7

“Judged by a higher standard, I have serious reservations. Remember, I wouldn’t go to the trouble of giving you this feedback if I didn’t think, based on what I’ve read in your letter, that you are capable of meeting the higher standard I mentioned.” +2 pages of critical feedback

Chart2





White Students

3.47

2.31

2.73

2.01

2.04

2.47

Fig 1 data


181417

31979

51458

76318

910619

111247

13532

15433


Fig 1 (scatterplot)

111

333

555

777

999

111111

131313

151515

Bluegill

Pumpkinseed

Green Sunfish

Time (weeks)



8

14

17

19

7

9

14

5

8

6

3

18

10

6

19

12

4

7

5

3

2

4

3

3

Fig 2 data

YearPopulation Size

198010

198115

198225

198320

198440

1985100

1986180

1987210

1988270

1989285

1990295


Fig 2 (line graph)

1980

1981

1982

1983

1984

1985

1986

1987

1988

1989

1990

Population Size

Year

N (Population Size)


10

15

25

20

40

100

180

210

270

285

295

Fig 3 data



White Students2.312.012.470.350.27

0.280.28


Fig 3 data

0.350.350.260.26

0.350.350.270.27

0.280.280.280.28


White Students

Fig 3 (bar graph)





Nutrient



N = 50 lakes.

3.47

2.73

2.31

2.01


1

2

3

4

5

UnbufferedCriticism



+ Assurance


Scale range: 1-7

Chart2





White Students

3.47

2.31

2.73

2.01

2.04

2.47

Fig 1 data


181417

31979

51458

76318

910619

111247

13532

15433


Fig 1 (scatterplot)

111

333

555

777

999

111111

131313

151515

Bluegill

Pumpkinseed

Green Sunfish

Time (weeks)



8

14

17

19

7

9

14

5

8

6

3

18

10

6

19

12

4

7

5

3

2

4

3

3

Fig 2 data

YearPopulation Size

198010

198115

198225

198320

198440

1985100

1986180

1987210

1988270

1989285

1990295


Fig 2 (line graph)

1980

1981

1982

1983

1984

1985

1986

1987

1988

1989

1990

Population Size

Year

N (Population Size)


10

15

25

20

40

100

180

210

270

285

295

Fig 3 data



White Students2.312.012.470.350.27

0.280.28


Fig 3 data

0.350.350.260.26

0.350.350.270.27

0.280.280.280.28


White Students

Fig 3 (bar graph)





Nutrient



N = 50 lakes.

3.47

2.73

2.31

2.01

Building trust at key transitions

“Wise feedback” group:I’m giving you these comments because I have high standards and I know that you can meet them.

Control group:I’m giving you these comments so you have feedback on your essay.

PresenterPresentation NotesJust one interaction – assume it happens over and over and over – less mysterious why and how trust leads to accumulating consequences

Percentage of students revising their essay:

• White students, both conditions : 75%• African American students, normal criticism: 17%• African American students, wise criticism: 71%*

*p

Seven years later, percentage of students attending 4-year college:

• White students, both conditions 48%• African American students, normal criticism: 45%• African American students, wise criticism: 64%*

*p = .01

Source: Yeager et al. (2017), Child Development

Shoring up belonging

Shoring up belonging

Greg Walton, Stanford

Shannon Brady, Stanford

Hardships Reported by Students in College Diary Study

Hardships Reported by Students in College Diary Study

• My teacher returned my paper covered in red ink.• Everyone is going out without me, and they didn’t consider me

when making their plans• My teacher cancelled her meeting with me. • A peer didn’t email me back.• I haven't gone on any dates.• I wasn’t recognized at awards dinner.• Dumped by girlfriend.• My boyfriend didn't call. • I’m working on a paper that is due tomorrow and I have writers

block.• Found a dead mouse under a pile of my clothes.

Belonging intervention at college transition:Sample survey quotes presented to participants in

intervention condition:

After winter break, I realized that all my good friends were at home. I didn’t have friends like that at school. But I got involved in extra-curriculars, and met people who had common interests. I also got to know people in class who later became close friends. But this took time. And before I found my niche here, there were times when I felt quite lonely.

-Respondent # 77, senior, White female

I worried that I was different from other students . . . . Now it seems ironic. Everybody feels they are different freshman year from everybody else, when really in at least some ways we are all pretty similar.

-Respondent, #23, African American female

Survey [Intervention materials shown to participants]

504 ** Juniors and Seniors completed survey materials . Participants were randomly sampled from the population of all [school name] Juniors and Seniors. Percentages are accurate within +/- 4 percentage points …

Results were consistent across class year and across racial and gender groups.

During their freshman year, many if not most students worry about whether other people at [college name] accept them.

• 73% - 86% of upperclassmen reported that, during their freshman year, they:• “sometimes” or “frequently” worried whether other students would accept them in

the context of classes and coursework.…

• “sometimes” or “frequently” felt intimidated by professors.

But after their freshman year, most students come to feel confident that other people at [college name] accept them.

• 82% - 97% of upperclassmen reported that, since their freshman year• they are “confident” or “certain” that most other students accept them

,,,,• “sometimes” or “frequently” felt their comfort in the academic environment at ** has

improved “some” or “a lot.”

•

College grade point average by year

Intervention

Walton & Cohen (2011, Science)

PresenterPresentation NotesWhite students showed the same pattern throughout. In terms of raw means, the treatment reduced the race gap in GPA over this period by 63%. Pause Beyond the main effect, there are four interesting patterns here.First, in the control condition, minority students show a long slump in performance through sophomore and junior year. They may be experiencing a recursive cycle – or an exacerbation cycle – whereby feelings of non-belonging lead to negative interactions with peers and teachers, reifying these feelings of non-belonging and undermining performance over time.2. Second, the treatment seems to interrupt this pattern leading to a linear increase in performance.3. Third, White students are stable over time.4. And fourth, towards the end of their college careers, the control condition seems to recover. This kind of analysis illustrates one value of field research. If you only study these interventions in the laboratory or at a single point in time, you may only capture the initial stage of a process that unfolds over time

Academic Mentor?

Continued Post-grad?

0%

20%

40%

60%

80%

100%

EuropeanAmerican

AfricanAmerican

Condition: Χ2(1)= 4.67, p=0.03 Condition: Χ2(1)= 15.05, p=0.002

0%

20%

40%

60%

80%

100%

EuropeanAmerican

AfricanAmerican

PresenterPresentation NotesSimple effects: EA: p = 0.093AA: p = 0.006

11 items. Highly reliable scale. Belonging uncertainty (2 items)Meaningful/rewarding (2 items) Belonging (3 items)Satisfaction (2 items)Potential/current success (2 items)

Seven years later…

Seven years later…Sense of belonging in career

Seven years later…Sense of belonging in career

-0.8

-0.6

-0.4

-0.2

0

0.2

0.4

0.6

White Black

Com

posi

te

(sta

ndar

dize

d)

The benefits of social belonging interventions have been replicated with a variety of populations (Clarke & Aguilar; Yeager, Walton et al).

Percentage of doctoral students in engineering who feel they “belong” in their field (Clarke, Aguilar, et al.)

-Men, both conditions combined: 65%-Women, control: 39%

Percentage of doctoral students in engineering who feel they “belong” in their field (Clarke, Aguilar, et al.)

-Men, both conditions combined: 65%-Women, control: 39%-Women, social belonging condition: 64%*

*Treatment effect, p < .05

Urban Students Graduating High School:Full-Time College Enrollment

(Yeager, Walton, Duckworth, Cohen, Dweck, et al., 2016, PNAS)

61%55%

41%32% 33%

59%53%

49% 44% 43%

20%30%40%50%60%70%80%

1 2 3 4 5Semester After High School

Cohort 1: N= 333

ControlSocial belonging

51% 52% 40%

57% 54%48%

20%30%40%50%60%70%80%

1 2 3Semester After High School

Cohort 2: N=1,060

ControlSocial belonging

* * *

*

* Significant controlling for high school GPA and SAT

Social Integration in Fall(Optional Fall survey; n = 51)

62% 52% 66%91% 76% 77%0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

Using academicsupport services

Living on campus Extracurriculars

Perc

enta

ge R

epor

ting

Beha

vior

ControlBelonging

Treatment vs. control: p = .005Yeager, Walton…& Dweck (2016)

Sum score fully mediates treatment effect on full-time enrollment

Because my parents didn’t go to college, they weren’t always able to provide me the advice I needed. So it was sometimes hard to figure out which classes to take and what I wanted to do in the future. But there are other people who can provide that advice, and I learned that I needed to rely on my adviser more than other students.

Difference-education Condition

PresenterPresentation NotesFirst, here’s an excerpt of a story from a FG panelist (Matt) [read slide]You can see from this example, his story highlighted his social class background, linked that background to the obstacles he faced in college, and also revealed strategies for success taking that particular background into account.

Difference-education Condition

I went to a small private school, and it was great college prep. We got lots of one-on-one attention, so it was a big adjustment going into classes with 300 people. I felt less overwhelmed when I took the time to get to know other students in the class.

PresenterPresentation NotesNow, here’s an excerpt from Rachel, a CG speaker (answering the same question). After subtly mentioning her parents’ graduate-level degrees in a prior response, she said [READ SLIDE]. Like Matt, Rachel talked about her background, a challenge she faced, and how she learned to meet that challenge. This speaker’s story also highlighted how her background mattered, but the contrast in stories showed how different backgrounds could shape experience and what you need to be successful.

Across both = see how experience links with your social class background; not just connected to obstacles but also strengths and strategies for success

Standard Condition

Go to class, and pay attention. If you don’t understand something or have a hard time with the material, meet with your teaching assistant or professor during office hours.

PresenterPresentation NotesAnd last, here’s an example from the control condition – Matt, the first-generation speaker we heard from earlier, was asked… READ SLIDE. He said… READ SLIDE.Across both conditions, the panelists mentioned learning to understand and overcome a challenge that they faced (much of the content was quite similar). The main difference was that, in the difference-education condition, the panelists connected their stories to their backgrounds and revealed how students’ different backgrounds can shape what they experience in college.

3.16 3.403.46 3.51

2.5

2.7

2.9

3.1

3.3

3.5

3.7

Academic Diversity

Cum

ulat

ive

GPA

(End

of F

irst Y

ear)

First-Generation Continuing-Generation

Standard Difference-educationStephens, Hamedani, & Destin, 2014, Psychological Science Note: All analyses control for race, gender, high school GPA, SATs, & income. F(1,125) = 4.34, p = .04

Gap No Gap

First Year GPA

PresenterPresentation NotesHere, we are looking at a graph depicting students’ cumulative GPAs after completing their first year. On the Y AXIS… ON X AXIS…As expected, in the standard control condition, we see the typical achievement gap in students’ GPA. But in the difference-education condition—after a 1-hour program in which students learned a framework to understand how their backgrounds matter—we closed the achievement gap and first-generation students earned significantly higher GPAs.

Notes:Interestingly, FG students are far less likely to engage in these behaviors; in part, because they get the idea that they should be able to figure everything out on their own! The difference-education condition closed 63% of the social class performance gap of .30 GPA points compared to the standard control condition.Main effect of condition, driven by FGs.

Instilling a growth mindset

Carol Dweck,Stanford

David Yeager,U Texas, Austin

Joshua Aronson,New York University

A Mindset Workshop• Control Group: 8 sessions of study skills.

• Growth Mindset Group: 8 sessions of study skills + the growth mindset. – Discuss scientific article on brain’s capacity to grow (“learning makes

you smarter”)– Discuss how mistakes are necessary for learning, how even “geniuses”

weren’t always geniuses (“10K hours of practice to become an expert”)– Write letters and discuss personal experiences that reinforce the growth

mindset

Math Grades(Blackwell, Trzesniewski, & Dweck)

2.2

2.3

2.4

2.5

2.6

2.7

2.8

ControlGrowth mindset

Percent Showing Increased Motivation

9

27

0

5

10

15

20

25

30

Control Growth

Instilling growth mindset

• Praising effort and strategy rather than ability

• Valuing mistakes

• Avoiding essentialist language

Affirming the self

Julio Garcia, Stanford Valerie Purdie Vaughns, Columbia

Jonathan Cook, Penn State David Sherman, UCSB Kevin Binning, U. Pittsburgh

Parker Goyer, Stanford

One of the lessons that I grew up with was to always stay true to yourself and never let what somebody else says distract you from your goals. And so when I hear about negative and false attacks, I really don’t invest any energy in them, because I know who I am.

-Michelle Obama, 2008

[VALUES AFFIRMATION EXERCISES]

WHAT ARE YOUR PERSONAL VALUES?

Remember that there are no right or wrong answers to any of these questions.Please read this list carefully. Then circle the TWO or THREE things that are MOST important to you.

Enjoying Sports

Being Good at Art

Being Creative

Being Independent

Living in the Moment (or Enjoying Today)

Belonging to a Social Group

(such as your community, racial group, or school club)

Listening to Music or Playing Music

Following Politics or Government

Being with Friends or Family

Being Religious

Having a Sense of Humor

[VALUES AFFIRMATION EXERCISE]

Look at the things you picked as MOST important to you on PAGE 1. Think

about times when these things were very important to you. Write a few

sentences describing why these things are important to you. Focus on your thoughts and feelings, and don’t worry about spelling or how well written it is.

___________________________________________________________________

___________________________________________________________________

___________________________________________________________________

___________________________________________________________________

___________________________________________________________________

___________________________________________________________________

Examples of affirmationsTeenagers

To me, independence is important because your parents won’t always be there to baby you. You have to live your life.

My friends and family are most important to me when I have a difficult situation. My friends give me companionship and courage. My family gives me love and understanding.

Adults

My religious values are the foundation of my life. I have always had a strong faith which has taught me to love others and led me to be a better person. My faith has taught me to be grateful, to trust that everything will be fine, and to enjoy every day as if it were the last.

For me, sense of humor is the most important thing. Every time someone makes me laugh it gives me comfort and happiness.

Values Affirmation in Middle School

Percentage of 7th graders receiving a D or F in thecourse where intervention given:

• White students, both conditions: 6%• Black students, control: 20%• Black students, affirmation: 9%*

Cohen et al., 2006, 2009, Science

One year after intervention, percentage of low-performing students assigned to remedial track:

• White students, both conditions 3%• African American students, control 18%• African American students, affirmation 5%*

*Comparison with control, p < 0.01

Cohen et al., 2009, Science


Seven years after intervention, percentage of students in college:

• White students, control and affirmation 80%• African American students, control 76%• African American students, affirmed 87%*

*Comparison with control, p = 0.01

• dddGoyer, et al. (2017) Proceedings of National Academy of Sciences



• White students, both conditions 80%• African American students, control 76%• African American students, affirmed 87%*


Goyer, et al. (2017) Proceedings of National Academy of Sciences


Seven years after intervention, percentage of students inselective college:




Replication with Disadvantaged Latino Middle SchoolersOver three years

(Goyer, Binning, Sherman, Purdie-Vaughns, Cook, Cohen, et al. 2012)

0%

10%

20%

30%

40%

50%

60%

70%

80%

Latinos in Remedial clinic Latinos in college-readinesselective

ControlAffirmation


Replication with Disadvantaged Latino Middle SchoolersOver three years(Goyer et al. 2012)

0%

10%

20%

30%

40%

50%

60%

70%

80%

Latinos in Remedial clinic Latinos in college-readinesselective

ControlAffirmation


Catalysts for change

• Trust

• Belonging

• Growth mindset

• Self affirmation

Affirmation:Applications in other contexts

• First year teachers (Brady et al.)• Financial aid uptake (Fotuhi et al.)• Online education (Kizilcek et al.)• Cognitive performance of the poor (Schwalbe et al.)• Medical compliance and health (Havranak et al.)

Scaling psychological interventions and wisdom

Disseminating interventions through technology: Using a moble application (app) to increase financial aid uptake

Percentage of poor students receiving financial aid (Fotuhi et al.)

• Control 39%• Affirmation 22%• App Reminder 58%• Affirmation + App Reminder 78%

Omid FotuhiOStanford

Scaling up:Interventions in Online Learning Contexts

Rene Kizilcek, Stanford

Interventions Close the Global GapA value relevance affirmation and a social-belonging intervention raise performance in less developed countries

Kizilcec, Saltarelli, Reich, & Cohen, in press, Science

N = 2,286 (16% in LDCs)

Scaling psychological intervention(Greg Walton, Dave Paunesku, David Yeager)

• PERTS (https://www.perts.net/)

• Community Transition Collaborative(collegetransitioncollaborative.org)

Scaling psychological wisdom

• Working with teachers (Shannon Brady, Stanford; Jason Okanafu, Berkeley)

• Working with tutors (Mark Lepper)

• Working with institutions (Shannon Brady, Stanford)

Key purposes of academic probation letter: Lost in communication (Brady et al.)

6 point scales from 1 = “not a purpose” to 6 = “a very important purpose”

AdminM

StudentM Diff

Connect students with helpful resources 5.57 3.96 1.61

Warn students that they are not meeting academic expectations 5.57 5.39 .18Let students know that staff/faculty are concerned and care about

them 5.35 3.29 2.06

Serve as a "wake-up call" 5.19 5.14 .05

Motivate students 4.97 3.84 1.13

Comply with federal and state regulations 4.33 4.16 .17

Identify students who can’t be successful 2.67 3.37 -0.7

Punish students 1.71 3.43 -1.72

PresenterPresentation Notes**Note: The “student” ratings were made by RAs looking at letters from a subset of the universities the administrators were from. (235 administrators; 32 letters, each rated by 2-3 RAs in this way.) In the future, we hope to more directly pair administrator and student ratings of letters, but we haven’t had capacity to do that yet.

Message intended is not message received.

If we can change the LETTER then we might be able to shift these downstream outcomes.So that’s what we did. We drew on social psychological theory, previous intervention work, and feedback from students to try to design a letter that would KEEP ALL OF THE CORE, ESSENTIAL content but be less likely to elicit high levels of shame and stigma.

Survey: College students who had been on academic probation said...

PresenterPresentation NotesThis word cloud was created during the coding process by selecting any poignant emotion driven wording regarding probation and creating a list. This list of words was then compiled and slightly altered. The alterations were done to make any verb tenses one single word, such as “stress” and “stressed”. Verbs and adjectives, on the other hand, were left separate in order to keep them closer to their meaning. The size of the word below is created by how many times the certain word is repeated. Since the word cloud generator was unable to keep phrases together, phrases such as “learning experience” and “let down” which only occurred once, were omitted. Phrases such as “insult to injury” were separated into “insult” and “injury”.

Intervention: Revise the letter• Frame probation as process not a label

– Placement on academic Probation…– The process for academic probation…

• Communicate “you’re not the only one”– You should also know that you are not alone in experiencing these difficulties.

• Acknowledge many reasons– Whatever difficulties– There are many reasons students enter the academic probation process. These reasons can

include personal, financial, health, family, or other issues.

• Offer hope for returning to good standing– By working with their advisors, many [students on probation] leave the process and continue

a successful career at [School].

• Offer testimonials from previous students on probation

PresenterPresentation NotesAlso appended “student perspectives on the probation process” which reinforced many of these ideas

THIRD WAY

Percentage of students still enrolled at the university (Brady et al.)

• Standard letter: 48%


• Standard letter: 48%• Revised letter: 79%

Affirming teacher affects the classroom (Brady et al)

1

2

3

4

5

Classroom Rigor Student Growth

Aver

age

Rat

ing

by M

ento

r Tea

cher

(5 p

oint

sca

le, h

ighe

r = m

ore)

ControlAffirmation

_________p = 0.03, d = .43

_________p = 0.04, d = .39

Error bars: 1 SE

PresenterPresentation NotesRated on 5-point rubric at end of first year by mentor teacher. Including random effects for mentor teacher increases the significance of the key DVs. Error bars are 1 SE.

Big Picture

THANK YOU!

Research Team, Collaborators, and Funders

AcknowledgmentsTeachers, administrators, and students at our school sites; Eric Bettinger and Demetra Logan; Lior Goldin

Faculty and student collaboratorsJulio Garcia; Valerie Purdie-VaughnsJonathan Cook; Greg WaltonDavid Sherman; David YeagerClaude Steele; Lee RossChristine Logel; Rachel SumnerArielle Silverman; Arghavan SallesKody Manke

Research Project TeamNancy ApfelStephanie ReevesAmy PetermannEric GomezWonhee LiuSuzanne Taborsky-BarbaPatti BrzustoskiAllison MasterCourtney BearnsNatalie GolaszewskiSarah TomassettiSarah WertMarie Scully

Research AssistantsSara Abiola; Stephanie GrayMichael Gray; Annie HsuLillian Hsu; Alyssa Mohammed Christopher Simmons; Eden Davis; James Lanham

Research ConsultantsEdward ZiglerDonald GreenEdmund GordonJoseph Mahoney

Funding SourcesNational Science FoundationSpencer FoundationW.T. Grant FoundationAmerican Psychological AssociationInstitute for Social and Policy StudiesNational Institute of Mental Health Nellie Mae Education Foundation School of Arts & Sciences, Yale University Society for the Psychol. Study of Social Issues

The person and the situation interact through time, changing both.

• Affirming students affects how they process later stressors.

Shannon Brady, StanfordJ. Ed Psych (2016)

-0.12

-0.10

-0.08

-0.06

-0.04

-0.02

0.00

0.02

0.04

0.06

0.08

0.10

0.12

Cont

rol

Belo

ngin

g

Cultu

re

Wise

feed

back

***CTR

A ge

ne e

xpre

ssio

n (lo

g 2m

RN

A)

Psychological intervention in first year of college can affect gene expression 4 years later(Brady, Cole, Walton, Ospina…, Cohen)

CTRA

Sympathetic Nervous System

Threat

Refra

med

Crit

icism

PresenterPresentation NotesData are change scores (T2 – T1) for a “CTRA contrast score” computed over measured levels of RNA produced by 41 genes involved in inflammation (up-regulated, weighted +1) and antiviral response (down-regulated, weighted -1). CTRA is a risk factor for chronic diseases (heart attacks, cancer, neurodegenerative diseases) and infections (due to poor antiviral response). As such, the general trend toward increase over time in most conditions is “bad” and the exceptional reduction for the wise feedback condition is “good”. *** indicates p < .001 difference from change in control condition as estimated from a mixed effect model treating the genes as repeated measures and additionally controlling for age, sex, race, BMI, any history of smoking, and history of heavy alcohol consumption.

-0.12

-0.10

-0.08

-0.06

-0.04

-0.02

0.00

0.02

0.04

0.06

0.08

0.10

0.12

Cont

rol

Belo

ngin

g

Cultu

re

Wise

feed

back

***CTR

A ge

ne e

xpre

ssio

n (lo

g 2m

RN

A c

ontra

st)


CTRA


Threat

Refra

med

Crit

icism


-0.12

-0.10

-0.08

-0.06

-0.04

-0.02

0.00

0.02

0.04

0.06

0.08

0.10

0.12

Cont

rol

Belo

ngin

g

Cultu

re

Wise

feed

back

***CTR

A ge

ne e

xpre

ssio

n (lo

g 2m

RN

A c

ontra

st)


CTRA


Threat

Posit

ive

Fram

e on

Cr

itici

sm in

Sch

ool


b = 0.21 letter grades p = .016

Affirming more students improves the classroom as a whole.

142

Powers et al., 2016Psychological Science

Joe Powers, Stanford

PresenterPresentation Notes{__ build up with how often these things can be overlooked}{really drive the excitement of you points home here, excitement is in the setting up of the question and the build up. Results confirm what is longed for.}Above and beyond the the previously reported gains of affirmation, individual student also showed increased end of fall grades from being in classrooms with greater proportions of affirmed African Americans. In the diagram above the y-axis is showing the students’ end-of-fall grade in the intervention-targeted class. The x-axis is showing the treatment density of each student’s classroom. 0 represents classrooms with even numbers of African-Americans in the intervention and control groups. So in a classroom where all students participanted, +2 represents classes with 2 more African Americans in intervention than in control. -2 represents classes with two fewer African Americans in intervention than in control. Students were seeing their grades improve by about 1/5 of a letter grade for every one unit increase in the treatment density of their classroom. Just a few more students protected from stereotype threat is raising the average grades of all classmates from a B to a B+. This classroom-level effect was statistically independent of the individual-level gains from treatment. Regardless of treatment condition and regardless of race, students still benefitted just from being in classrooms with larger proportions of affirmed African Americans.However, among all races and treatment conditions, students with a history of low prior performance benefit the most [NEXT SLIDE]

b = 0.21 letter grades p = .016

Affirming more students improves the classroom as a whole.

143

Powers et al., 2016Psychological Science

Joe Powers, Stanford

PresenterPresentation Notes{__ build up with how often these things can be overlooked}{really drive the excitement of you points home here, excitement is in the setting up of the question and the build up. Results confirm what is longed for.}Above and beyond the the previously reported gains of affirmation, individual student also showed increased end of fall grades from being in classrooms with greater proportions of affirmed African Americans. In the diagram above the y-axis is showing the students’ end-of-fall grade in the intervention-targeted class. The x-axis is showing the treatment density of each student’s classroom. 0 represents classrooms with even numbers of African-Americans in the intervention and control groups. So in a classroom where all students participanted, +2 represents classes with 2 more African Americans in intervention than in control. -2 represents classes with two fewer African Americans in intervention than in control. Students were seeing their grades improve by about 1/5 of a letter grade for every one unit increase in the treatment density of their classroom. Just a few more students protected from stereotype threat is raising the average grades of all classmates from a B to a B+. This classroom-level effect was statistically independent of the individual-level gains from treatment. Regardless of treatment condition and regardless of race, students still benefitted just from being in classrooms with larger proportions of affirmed African Americans.However, among all races and treatment conditions, students with a history of low prior performance benefit the most [NEXT SLIDE]

Affirming teacher affects the classroom (Brady et al)

1

2

3

4

5

Classroom Rigor Student Growth

Aver

age

Rat

ing

by M

ento

r Tea

cher

(5 p

oint

sca

le, h

ighe

r = m

ore)

ControlAffirmation

_________p = 0.03, d = .43

_________p = 0.04, d = .39

Error bars: 1 SE

PresenterPresentation NotesRated on 5-point rubric at end of first year by mentor teacher. Including random effects for mentor teacher increases the significance of the key DVs. Error bars are 1 SE.

Time and timing moderate the effects of psychological processes and the interventions that trigger them.

Timing matters:Affirming earlier in middle school leads to

broader benefits (Binning, Cook, et al.)

Timing matters:Affirming earlier in middle school leads to

broader benefits (Binning, Cook, et al.)

1.67

2.00

2.33

2.67

3.00

3.33

1 2 3 4 5 6 7 8 9 10 11 12

6th Grade 7th Grade 8th Grade

Raw GPA Over 12 Quarters

Control Affirmation

Do I belong?

Am I valued?

Does this have meaning?

Do I have something to learn?

Can I trust?

Am I safe?

Creating the right context:Combining psychological intervention with contextual channel:


• Control 39%• Affirmation 22%• Reminders 58%• Reminders + Affirmation 78%


Creating the right context: Combining psychological intervention with contextual channel:


• Control 39%• Affirmation 22%• Reminders 58%• Reminders + Affirmation 78%


Key purposes of academic probation letter: Lost in communication (Brady et al.)

6 point scales from 1 = “not a purpose” to 6 = “a very important purpose”

AdminM

StudentM Diff

Connect students with helpful resources 5.57 3.96 1.61

Warn students that they are not meeting academic expectations 5.57 5.39 .18Let students know that staff/faculty are concerned and care about

them 5.35 3.29 2.06

Serve as a "wake-up call" 5.19 5.14 .05

Motivate students 4.97 3.84 1.13

Comply with federal and state regulations 4.33 4.16 .17

Identify students who can’t be successful 2.67 3.37 -0.7

Punish students 1.71 3.43 -1.72

PresenterPresentation Notes**Note: The “student” ratings were made by RAs looking at letters from a subset of the universities the administrators were from. (235 administrators; 32 letters, each rated by 2-3 RAs in this way.) In the future, we hope to more directly pair administrator and student ratings of letters, but we haven’t had capacity to do that yet.

Message intended is not message received.

If we can change the LETTER then we might be able to shift these downstream outcomes.So that’s what we did. We drew on social psychological theory, previous intervention work, and feedback from students to try to design a letter that would KEEP ALL OF THE CORE, ESSENTIAL content but be less likely to elicit high levels of shame and stigma.

Survey: College students who had been on academic probation said...

PresenterPresentation NotesThis word cloud was created during the coding process by selecting any poignant emotion driven wording regarding probation and creating a list. This list of words was then compiled and slightly altered. The alterations were done to make any verb tenses one single word, such as “stress” and “stressed”. Verbs and adjectives, on the other hand, were left separate in order to keep them closer to their meaning. The size of the word below is created by how many times the certain word is repeated. Since the word cloud generator was unable to keep phrases together, phrases such as “learning experience” and “let down” which only occurred once, were omitted. Phrases such as “insult to injury” were separated into “insult” and “injury”.

Intervention: Revise the letter• Frame probation as process not a label

– Placement on academic Probation…– The process for academic probation…

• Communicate “you’re not the only one”– You should also know that you are not alone in experiencing these difficulties.

• Acknowledge many reasons– Whatever difficulties– There are many reasons students enter the academic probation process. These reasons can

include personal, financial, health, family, or other issues.

• Offer hope for returning to good standing– By working with their advisors, many [students on probation] leave the process and continue

a successful career at [School].

• Offer testimonials from previous students on probation

PresenterPresentation NotesAlso appended “student perspectives on the probation process” which reinforced many of these ideas

THIRD WAY

Scaling up:Interventions in Online Learning Contexts

Rene Kizilcek, Stanford

Cultural Differences in MCII’s EffectsAn 8-minute self-regulation intervention raises course completion in individualist but not in collectivist cultures

Kizilcec & Cohen, R&R



N = 2,286 (16% in LDCs)

Experiment 1: Subsequent Enrollment

N = 2,286



N = 1,165 (7.4% in LDCs)

Big Picture

Research Team, Collaborators, and Funders

AcknowledgmentsTeachers, administrators, and students at our school sites; Eric Bettinger and Demetra Logan

Faculty and student collaboratorsJulio Garcia; Valerie Purdie-VaughnsJonathan Cook; Greg WaltonDavid Sherman; David YeagerClaude Steele; Lee RossChristine Logel; Rachel SumnerArielle Silverman; Arghavan SallesKody Manke

Research Project TeamNancy ApfelStephanie ReevesAmy PetermannEric GomezWonhee LiuSuzanne Taborsky-BarbaPatti BrzustoskiAllison MasterCourtney BearnsNatalie GolaszewskiSarah TomassettiSarah WertMarie Scully

Research AssistantsSara Abiola; Stephanie GrayMichael Gray; Annie HsuLillian Hsu; Alyssa Mohammed Christopher Simmons; Eden Davis; James Lanham

Research ConsultantsEdward ZiglerDonald GreenEdmund GordonJoseph Mahoney

Funding SourcesNational Science FoundationSpencer FoundationW.T. Grant FoundationAmerican Psychological AssociationInstitute for Social and Policy StudiesNational Institute of Mental Health Nellie Mae Education Foundation School of Arts & Sciences, Yale University Society for the Psychol. Study of Social Issues

Barriers in editorial and institutional gatekeeping

Psychological System

Behavior

Cohen & Sherman, Annual Review, 2014

Behavior



Behavior

Social System



Behavior

Social System



Slide Number 1Powerful but invisible:�Social psychological processes shaping student successSlide Number 3Slide Number 4Slide Number 5Slide Number 6Slide Number 7Slide Number 8Feeling like an outsider:�A common predicamentFeeling like an outsider:�A common predicamentFeeling like an outsider:�A common predicamentFeeling like an outsider:�A common predicamentFeeling like an outsider:�A common predicamentFeeling like an outsider:�A common predicamentThe same classroom can be experienced more differently than different classrooms.��The same classroom can be experienced more differently than different classrooms. �The same classroom can be experienced more differently than different classrooms.�Slide Number 18Slide Number 19Slide Number 20Slide Number 21Slide Number 22Slide Number 23Stereotype threat experiment:�When White men can’t jumpStereotype threat experiment:�When White men can’t jumpTest Performance�(Number Correct)Test Performance�(Number Correct)Math Test Performance Among Men and Women�(Percent Correct)Math Test Performance Among Men and Women�(Percent Correct)Slide Number 30Slide Number 31Slide Number 32Slide Number 33Slide Number 34Slide Number 35Slide Number 36Changing the social environmentChanging the social environmentSlide Number 39Slide Number 40Slide Number 41Counteracting the costs of vigilanceCounteracting the costs of vigilanceCounteracting the costs of vigilanceSlide Number 45Slide Number 46Slide Number 47Catalysts for changeCreating turning points:�When social influence lastsCreating turning points:�When social influence lastsSlide Number 51Slide Number 52Perceived Instructor BiasPerceived Instructor BiasPerceived Instructor BiasPerceived Instructor BiasPerceived Instructor BiasPerceived Instructor BiasPerceived Instructor BiasSlide Number 60Slide Number 61Slide Number 62Slide Number 63Slide Number 64��Seven years later, percentage of students attending 4-year college:��Slide Number 66��Shoring up belonging��Hardships Reported by Students in College Diary StudyHardships Reported by Students in College Diary StudyBelonging intervention at college transition:�Sample survey quotes presented to participants in intervention condition:Slide Number 71Slide Number 72Slide Number 73Slide Number 74Slide Number 75Slide Number 76�Seven years later…��Seven years later…�Sense of belonging in career��Seven years later…�Sense of belonging in career�Slide Number 80Slide Number 81Slide Number 82Urban Students Graduating High School:�Full-Time College Enrollment�(Yeager, Walton, Duckworth, Cohen, Dweck, et al., 2016, PNAS)Social Integration in Fall�(Optional Fall survey; n = 51)Slide Number 85Slide Number 86Slide Number 87Slide Number 88Slide Number 89A Mindset WorkshopSlide Number 91Slide Number 92Math Grades�(Blackwell, Trzesniewski, & Dweck) Percent Showing �Increased MotivationInstilling growth mindsetAffirming the selfSlide Number 97Slide Number 98Slide Number 99Examples of affirmationsSlide Number 101Values Affirmation in Middle SchoolSlide Number 103Values Affirmation in Middle SchoolValues Affirmation in Middle SchoolValues Affirmation in Middle SchoolValues Affirmation in Middle SchoolReplication with Disadvantaged Latino Middle Schoolers�Over three years�(Goyer, Binning, Sherman, Purdie-Vaughns, Cook, Cohen, et al. 2012)Replication with Disadvantaged Latino Middle Schoolers�Over three years�(Goyer et al. 2012)Slide Number 110Slide Number 111��Slide Number 113Slide Number 114Slide Number 115Slide Number 116Slide Number 117Catalysts for changeAffirmation:�Applications in other contextsSlide Number 120Disseminating interventions through technology: �Using a moble application (app) to increase financial aid uptake�Percentage of poor students receiving financial aid (Fotuhi et al.)Scaling up:�Interventions in Online Learning ContextsInterventions Close the Global GapScaling psychological intervention�(Greg Walton, Dave Paunesku, David Yeager)Scaling psychological wisdomScaling psychological wisdomScaling psychological wisdomKey purposes of academic probation letter: �Lost in communication �(Brady et al.)Survey: College students who had been on academic probation said...Intervention: Revise the letterPercentage of students still enrolled at the university (Brady et al.)Percentage of students still enrolled at the university (Brady et al.)Affirming teacher affects the classroom (Brady et al)Slide Number 134Slide Number 135Research Team, Collaborators, and FundersSlide Number 137Slide Number 138Slide Number 139Slide Number 140Slide Number 141Affirming more students improves the classroom as a whole.Affirming more students improves the classroom as a whole.Affirming teacher affects the classroom (Brady et al)Slide Number 145Timing matters:�Affirming earlier in middle school leads to �broader benefits (Binning, Cook, et al.)Timing matters:�Affirming earlier in middle school leads to �broader benefits (Binning, Cook, et al.)Slide Number 148Percentage of students still enrolled at the university (Brady et al.)Percentage of students still enrolled at the university (Brady et al.)Creating the right context:�Combining psychological intervention with contextual channel:�Percentage of poor students receiving financial aid (Fotuhi et al.)Creating the right context: �Combining psychological intervention with contextual channel:�Percentage of poor students receiving financial aid (Fotuhi et al.)Key purposes of academic probation letter: �Lost in communication �(Brady et al.)Survey: College students who had been on academic probation said...Intervention: Revise the letterPercentage of students still enrolled at the university (Brady et al.)Percentage of students still enrolled at the university (Brady et al.)Scaling up:�Interventions in Online Learning ContextsCultural Differences in MCII’s EffectsInterventions Close the Global GapExperiment 1: Subsequent EnrollmentInterventions Close the Global GapSlide Number 163Research Team, Collaborators, and FundersBarriers in editorial and �institutional gatekeepingSlide Number 166Slide Number 167Slide Number 168Slide Number 169

powerful but invisible: social psychological processes ... · coach wooden and me. this need to be...

Documents