Blackwell (2007) Implicit Theories of Intelligence Study

Blackwell et al. (2007) study tested whether believing intelligence can grow helps students in school. It reports two separate studies of American seventh graders. In Study 1, students who believed intelligence can be developed saw their maths grades rise over two years. In Study 2, a growth-mindset lesson stopped a fall in the grades of lower-achieving students.

Key Takeaways

Aim

  • To test whether a growth mindset (believing intelligence can grow with effort) is linked to maths grades.
  • A fixed mindset is believing it cannot. Study 2 also tested whether teaching a growth mindset helps.

Method

  • Study 1 followed the same 373 seventh graders (Year 8, aged 12 to 13) for two years in a real school. Following people over time is a longitudinal study.
  • It was correlational: it measured whether mindset (a questionnaire score, higher for more growth mindset) and maths grades were linked, and changed nothing.
  • Mindset was the predictor variable (used to predict). Maths grades were the outcome variable (predicted).
  • Study 2 was an experiment in another New York school with 91 lower-achieving seventh graders. Whole advisory groups (like tutor groups) were randomly put in a growth-mindset group (48) or a control group (43).
  • Both groups had eight weekly workshops, but only the growth-mindset group learned that intelligence can grow. Different students were in each group (independent measures).
  • The independent variable (what researchers changed) was the group. The dependent variables (what they measured) were mindset change, teacher reports of motivation, and maths grades.

Results

  • Study 1: at the start, mindset was not linked to maths scores. Over two years, growth-mindset students’ grades rose and fixed-mindset students’ grades stayed flat.
  • Study 1: growth-mindset students had stronger learning goals (wanting to improve), valued effort more, and blamed failure less on lack of ability.
  • Study 2: the growth-mindset group’s mindset score rose (4.36 to 4.95). The control group’s barely moved (4.62 to 4.68).
  • Study 2: a teacher who did not know the groups named 27% of the growth-mindset group as more motivated, against 9% of the control group.
  • Study 2: grades fell for the whole sample. The fall stopped in the growth-mindset group and carried on in the control group.

Conclusion

  • A growth mindset is linked to better maths grades over the move to junior high (the first years of secondary school). The authors link this to stronger motivation (Study 1).
  • Teaching a growth mindset can improve motivation and stop grades falling (Study 2).

Strengths

  • Ecological validity: both studies took place in real schools with real term grades, so the results fit real life. (Ecological validity means how well results apply to everyday life.)
  • Longitudinal design: Study 1 followed the same students for two years, so it shows real change over time.
  • Random allocation and a close control group: in Study 2 chance decided who went in which group, and the control workshop was very similar.
  • Reliability: students gave similar mindset scores when they did the questionnaire twice, two weeks apart. (Reliability means getting consistent results.)
  • Blind measurement: the teacher did not know who was in which group, so the reports were less likely to be biased.

Limitations

  • Cultural bias: both samples came from single schools in New York City, so the findings may not apply to students elsewhere.
  • Age bias: everyone was in seventh grade, so the findings may not apply to older or younger students.
  • Construct validity: only maths grades measured achievement, so the study may not show achievement in other subjects. (Construct validity means measuring what the study claims to measure.)
  • Social desirability: mindset was measured by questionnaire, so students may have answered the way they thought they should.
  • Cause and effect: Study 1 did not change mindset, so something else, such as a supportive teacher, could explain both mindset and grades.
  • Individual differences: Study 2 used different students in each group, so differences between them could have affected the results.

Ethical Issues

  • Informed consent and deception: the paper does not say students were told one workshop taught a growth mindset. Way of dealing with it: debrief them afterwards. (A debrief is a full explanation at the end.)
  • Right to withdraw: taking part was voluntary, and in Study 1 students could stop at any time. Parents and students gave consent.
  • Fairness to the control group: the controls missed the growth-mindset lesson. Way of dealing with it: give them the training afterwards.
  • Protection from harm and confidentiality: harm was unlikely, but grades are personal. Way of dealing with it: keep names private and use numbers.

Exam Use

  • Know which study the question asks about. Examiners report that students lose marks by mixing up Study 1 and Study 2.
  • A good description of Study 1 needs the procedure and the findings. Without both, the top mark is lost.
  • A criticism needs a reason tied to the study. “The sample is small” does not earn a mark on its own.
  • Blackwell can be used in the 13-mark question, the longest on the paper. You say how far you agree with a statement and back it with two studies.

Background

In the USA, students move from elementary school (like primary school) to junior high (the first years of secondary school) at about age 12 or 13. The move is hard. Many students see their grades fall, and some never recover.

Why do some students cope better? The study looked at their beliefs about intelligence. OCR calls these beliefs mindsets.

  • Fixed mindset (an entity theory): intelligence is fixed. You have a certain amount and cannot do much to change it.
  • Growth mindset (an incremental theory): intelligence can be developed through effort and learning.

The researchers expected the two mindsets to lead to four different habits.

  • Learning goals: growth-mindset students want to increase their ability. Fixed-mindset students want to prove it.
  • Effort beliefs: growth-mindset students think effort is useful. Fixed-mindset students think it is pointless.
  • Helpless attributions (reasons given for failing): growth-mindset students blame low effort. Fixed-mindset students blame low ability.
  • Strategies: after a setback, growth-mindset students try harder or change method. Fixed-mindset students give up.

Aim

The aim was to see whether students’ beliefs about intelligence are linked to their achievement. It also tested whether a set of lessons (an intervention) could change this link.

There were two studies. Each has its own hypothesis, a prediction the study can test.

  • Study 1: there will be a link between teenagers’ beliefs about intelligence and their maths grades.
  • Study 2: a lesson telling students that intelligence can be developed will improve their motivation and, later, their maths grades.

Study 1 or Study 2?

Examiners say students often mix up the two studies. The table keeps them apart.

Study 1Study 2
Students373 seventh graders, moderately high-achieving99 lower-achieving seventh graders, 91 in the intervention
MethodLongitudinal correlational field studyIntervention study with a control group, independent measures
What the researchers didMeasured mindset once, then followed grades for two yearsTaught one group a growth mindset and compared it with a control group
Main findingGrowth-mindset students’ grades rose and fixed-mindset students’ stayed flatThe fall in grades stopped in the growth-mindset group only

OCR has set a question on Study 1 by name: describe it, or give two criticisms of it. Learn both studies, and read the question to see which one it wants.

Method

Design and Variables

The paper reported two separate studies in two different New York schools.

  • Study 1: a longitudinal study, because the same students were followed over time. A field study, because it was in a real school. Correlational, because it measured whether two things are linked.
  • Study 1 variables: nothing was changed, so there is no independent variable. The two co-variables were mindset and maths grades. Mindset was the predictor variable and maths grades were the outcome variable.
  • Study 2: an intervention study: an experiment in a real school with a growth-mindset group and a control group. OCR’s own label is a correlational field study with an experimental section.
  • Study 2 design: independent measures, so different students were in each group.
  • Study 2 variables: the independent variable was the group (growth-mindset workshop or control). The dependent variables were change in mindset, the teacher’s reports of classroom motivation, and maths grades.

In Study 2, chance decided who went in which group.

The school had already put students into advisory groups of 12 to 14 (like tutor groups). Each whole group was randomly given to the growth-mindset condition or the control condition.

Sample

  • Study 1: 373 seventh graders (198 girls and 175 boys) at one public school in New York City. They were moderately high-achieving in sixth-grade maths (sixth grade is Year 7).
  • Study 1 make-up: an ethnically mixed sample: 55% African American, 27% South Asian and 15% Hispanic.
  • Study 2: 99 seventh graders (49 girls and 50 boys) at a different New York public school. They were relatively low-achieving in maths.
  • Study 2 make-up: 79% were eligible for free school lunch, a sign of low family income.
  • From 99 to 91: 95 chose to join. Five could not attend regularly and were dropped. That left 48 and 43, similar in maths grades at the start.

Materials

  • The questionnaire (both studies): students rated statements such as “You have a certain amount of intelligence, and you really can’t do much to change it.”
  • The mindset score: it runs from 1 (fixed mindset) to 6 (growth mindset). A higher score means more growth mindset.
  • Other scales: it also measured learning goals and effort beliefs. Students imagined scoring 54 (an F, the lowest grade) on a quiz and said whether they would blame their ability or work harder.
  • Maths measures (Study 1): a standardised citywide maths test in sixth grade, then term grades. A term grade combined tests, homework, a project and participation.
  • Study 2 materials: a reading called “You Can Grow Your Intelligence” for the growth-mindset group, a quiz for both groups, and the maths teacher’s report.

Procedure

Study 1:

  1. At the start of the autumn term of seventh grade, trained research assistants gave the questionnaire in a normal lesson.
  2. Sixth-grade maths scores were taken from school records. Maths grades were then collected at the end of every autumn and spring term for two years.
  3. Four groups of students, each entering in a different year, were followed. The project ran for five years, but each student was followed for two.

Study 2:

  1. Students filled in the questionnaire at the start of seventh grade.
  2. In the spring term they were invited to a voluntary eight-week workshop on the brain and study skills, with a certificate at the end.
  3. There were eight weekly 25-minute sessions, led by sixteen trained undergraduates in teams of two.
  4. Both groups had very similar sessions on the brain, on stereotyping and on study skills.
  5. In two sessions the growth-mindset group learned that learning changes the brain by making new connections, and that students are in charge of this. The control group had a memory lesson instead.
  6. At the end, everyone did a quiz. Three weeks later the questionnaire was repeated.
  7. The maths teacher wrote down which students had changed in motivation. The teacher did not know who was in which group.

Results

Study 1 Results (373 Students)

  • At the start: mindset was not linked to sixth-grade maths scores. So the beliefs were not just a result of earlier maths achievement.
  • The beliefs fitted together: growth-mindset students had stronger learning goals and valued effort more. They blamed their ability less after failure and planned to try harder.
  • Grades over time: a small link between mindset and grades appeared in junior high and grew over the two years. Growth-mindset students’ grades rose. Fixed-mindset students’ grades stayed flat.
  • A likely reason: motivational beliefs seemed to help explain the link between mindset and rising grades.

Study 2 Results (91 Students)

  • The quiz, general content: the quiz checked that the message had got across. Both groups did about as well on questions about the general workshop content (growth-mindset group 73.0%, control group 70.5%).
  • The quiz, growth-mindset content: the growth-mindset group scored 84.5% on these questions. The control group scored 53.9%. So only the growth-mindset group learned the message.
  • Mindset changed: the growth-mindset group’s mindset score rose from 4.36 to 4.95. The control group’s moved from 4.62 to 4.68, which is hardly any change.
  • Teacher’s reports: 17 students were named as improving in motivation or performance. 13 of them were in the growth-mindset group. That is 27% of the growth-mindset group and 9% of the control group.
  • Grades: grades fell across seventh grade for the whole sample. The groups did not differ before the workshops. After them, the fall stopped in the growth-mindset group and carried on in the control group.

The growth-mindset group’s grades edged up a little. They stopped falling, but they did not rise much.

The authors expected the lesson to help most those who started with a fixed mindset, and the data pointed that way. The groups were too small to be sure.

Conclusion

The paper draws four main conclusions.

  1. Growth mindset and motivation (Study 1): students with a growth mindset have stronger learning goals, value effort more and blame their ability less.
  2. Growth mindset and grades (Study 1): a growth mindset is linked to better maths achievement over the move to junior high. Before the move, fixed-mindset students seemed to be doing fine.
  3. Teaching a growth mindset (Study 2): teaching that intelligence can be developed improved motivation, in the teacher’s reports, and stopped the fall in maths grades within one term.
  4. A key factor: a student’s theory of intelligence is an important part of achievement motivation.

The first two conclusions come from Study 1 and the third from Study 2. The authors add that their effects are small.

Strengths

  • Ecological validity: both studies took place in real public schools with normal lessons and the students’ real term grades. So the results are likely to fit real life.
  • Longitudinal design (Study 1): following the same students for two years shows real change. Students are compared with themselves, so differences between individuals matter less.
  • A second school: Study 2, in a lower-achieving school, found a similar pattern between mindset, motivation and grades, though only just short of the usual cut-off. A finding in two schools is stronger.
  • Reliability: students scored very similarly when they did the mindset questionnaire twice, two weeks apart. So it measures mindset consistently.
  • Random allocation and a close control group (Study 2): chance decided which groups got which lesson, and both had a very similar workshop. So mainly the mindset lesson differed.
  • Blind measurement (Study 2): the teacher did not know who was in which group, and neither did the people who coded the comments. That reduces bias.
  • A check that the message got across (Study 2): the quiz showed that only the growth-mindset group learned the growth-mindset message.

Limitations

  • Cultural bias: both samples came from single schools in New York City, so they may not represent students elsewhere. The samples were ethnically mixed, so the point is place, not ethnic group.
  • Age bias: everyone was in seventh grade. The authors say the effect shows up during a hard move, so it may not apply to other ages.
  • Construct validity: only maths grades measured achievement. A term grade also mixes tests, homework, a project and participation. Mindset may work differently in English or art.
  • Social desirability: every belief was measured by students rating statements, so some may have answered the way they felt they should.
  • Cause and effect (Study 1): mindset was measured and not changed, so the link does not prove that mindset causes grades. A supportive teacher or family could raise both.
  • Individual differences (Study 2): different students were in each group, so differences between them, such as home life, could affect the results. Random groups and similar starting grades reduced this.
  • Extraneous variables (Study 2): other things may have changed the results, such as which student leaders taught a group. The growth-mindset group also had an extra talk about unfair labels such as “stupid”.
  • Small effects and a short follow-up: the effects were small, and students were followed for only a short time after the workshops.

Small Effects and a Short Follow-Up

The authors say their findings are based on small effects. In the exam, “the sample is small” is not enough by itself. Tie it to a result: the groups were too small to be sure that fixed-mindset students gained most.

The Nature and Nurture Debate

OCR names the nature and nurture debate as the criticism of Dweck’s mindset theory. This is a criticism of the theory, not of the study.

Nature is what we inherit. Nurture is what we learn from experience. The theory focuses too much on nurture, because it makes achievement depend on effort and praise. It ignores people who do well on innate intelligence alone.

Ethical Issues

Parents and students gave consent. Taking part was voluntary, and in Study 1 students could withdraw at any time without penalty. That is the right to withdraw.

  • Informed consent and deception: informed consent means agreeing after being told what is involved. The paper does not say students were told there were two versions. Way of dealing: debrief them afterwards.
  • Young participants: the students were children. Way of dealing: get consent from parents as well as students, which the paper reports.
  • Fairness to the control group: if the message helps, the controls missed out. Way of dealing: give them the training afterwards.
  • Protection from harm and confidentiality: the lessons were positive, so harm was unlikely. The paper does not say how grades were kept private. Way of dealing: use numbers, not names.

Exam Questions

What OCR Asks About This Study

In the exam you get marks for these question types, all taken from OCR’s past papers on this study.

  • Multiple choice (1 mark each): the method (a longitudinal study using questionnaires), what was measured (maths) and what students believed (that they could change their intelligence).
  • Describe Study 1 (4 marks): give the method, sample, questionnaire, grades and finding. With no procedure or findings, the top mark is 3. If you drift into Study 2, only what came first counts.
  • Explain two criticisms of Study 1 (4 marks): one mark for each criticism and one for each explanation in context. Credited: age bias, cultural bias, maths as a narrow measure, social desirability, extraneous variables.
  • Identify a bias in the sample (2 marks): cultural bias, linked to the study, such as only children from New York. “Only American children” earns one mark. Do not mix up city and country.
  • Suggest an improvement (2 marks): a limitation and a matching improvement, not about the sample. For example: measure progress in other subjects, or give the controls the training afterwards.
  • Evaluate the study (6 marks): points must be about the procedure. Evaluating only the findings is limited to 2 marks.
  • 13 marks: you say how far you agree with a statement, such as “Research using correlations is not useful”. You need Blackwell and a study from a different topic.

In a 13-mark answer, Blackwell cannot be paired with Piaget’s study, which is also from Development.

The three questions below are examples written in OCR’s style. They are not OCR’s own wording.

Practice Question 1: Describe Study 1 (4 marks)

Question [4 marks]: Describe Study 1 of Blackwell et al.’s research into fixed and growth mindsets.

Model answer: Study 1 was a longitudinal correlational field study. It took place at a public school in New York City. 373 seventh graders filled in a questionnaire at the start of seventh grade. It measured whether they had a fixed or a growth mindset.

Researchers then collected their maths grades every term for two years. At the start, mindset was not linked to maths scores. Over the two years, growth-mindset students’ grades rose, but fixed-mindset students’ grades stayed flat.

  • Mark scheme: Level 2 (3 to 4 marks) is a good description. Level 1 (1 to 2 marks) is a basic one. With no procedure or no findings the top mark is 3.
  • Watch out: do not describe Study 2 and do not evaluate.

Practice Question 2: Two Criticisms of Study 1 (4 marks)

Question: Explain two reasons why Study 1 of Blackwell et al.’s research can be criticised. [4 marks]

Model answer: One criticism is age bias. Everyone was in seventh grade, so the findings may not apply to older students.

Another criticism is construct validity. Only maths grades were used to measure achievement. So the study may not show how mindset affects other subjects.

  • Mark scheme: 2 marks for each criticism: 1 for naming it and 1 for explaining it in context. Any two credited criticisms score.

Practice Question 3: Evaluate Study 2 (6 marks)

Question: Evaluate Study 2 of Blackwell et al.’s research. [6 marks]

Model answer: One strength is the control group. Both groups had a very similar workshop, and chance decided which groups got which lesson. So mainly the mindset lesson differed.

A second strength is that the teacher did not know who was in which group, which reduces bias.

One weakness is individual differences. Different students were in each group, so differences between them could have affected the grades. An improvement would be to match students for earlier maths grades.

A second weakness is construct validity. Only maths grades were measured. An improvement would be to measure progress in other subjects too.

In the exam, a 6-mark evaluation is marked in three levels.

  • Top level (5 to 6): a thorough, relevant evaluation with some breadth and depth.
  • Middle level (3 to 4): a reasonably detailed evaluation, through breadth or depth.
  • Bottom level (1 to 2): a basic evaluation where at least one point is relevant.

As far as the saved papers show, the real 2021 question did not name Blackwell. So this practice question is only an example of the type.

Later Research

Yeager and colleagues tested a short online growth-mindset lesson on 12,490 ninth graders in a nationally representative sample of US schools. Students were randomly assigned to the lesson or a control. For lower-achieving students, grades rose by a small amount (Yeager et al., 2019).

Macnamara and Burgoyne combined 63 studies of growth-mindset interventions. They found a very small average effect on achievement. They concluded that apparent gains were probably due to weak study design, flaws in reporting and bias (Macnamara & Burgoyne, 2023).

So later evidence is mixed. The original study is valuable but small.

References

Blackwell, L. S., Trzesniewski, K. H., & Dweck, C. S. (2007). Implicit theories of intelligence predict achievement across an adolescent transition: A longitudinal study and an intervention. Child Development, 78(1), 246-263. https://doi.org/10.1111/j.1467-8624.2007.00995.x

Macnamara, B. N., & Burgoyne, A. P. (2023). Do growth mindset interventions impact students’ academic achievement? A systematic review and meta-analysis with recommendations for best practices. Psychological Bulletin, 149(3-4), 133-173. https://doi.org/10.1037/bul0000352

Yeager, D. S., Hanselman, P., Walton, G. M., Murray, J. S., Crosnoe, R., Muller, C., Tipton, E., Schneider, B., Hulleman, C. S., Hinojosa, C. P., Paunesku, D., Romero, C., Flint, K., Roberts, A., Trott, J., Iachan, R., Buontempo, J., Yang, S. M., Carvalho, C. M., … Dweck, C. S. (2019). A national experiment reveals where a growth mindset improves achievement. Nature, 573(7774), 364-369. https://doi.org/10.1038/s41586-019-1466-y

Olivia Guy-Evans, MSc

BSc (Hons) Psychology, MSc Psychology of Education

Associate Editor for Simply Psychology

Olivia Guy-Evans is a writer and associate editor for Simply Psychology, where she contributes accessible content on psychological topics. She is also an autistic PhD student at the University of Birmingham, researching autistic camouflaging in higher education.


Saul McLeod, PhD

Chartered Psychologist (CPsychol)

BSc (Hons) Psychology, MRes, PhD, University of Manchester

Saul McLeod, PhD, is a qualified psychology teacher with over 18 years of experience in further and higher education. He has been published in peer-reviewed journals, including the Journal of Clinical Psychology.