Fatskills
Practice. Master. Repeat.
Study Guide: AP Psychology – Statistics in Psychology (Descriptive, Inferential, p-value)
Source: https://www.fatskills.com/ap-psychology/chapter/ap-topic-guides-ap-psychology-statistics-in-psychology-descriptive-inferential-pvalue

AP Psychology – Statistics in Psychology (Descriptive, Inferential, p-value)

By Fatskills Exam Guides Team — the exam nerds behind 28,500+ quizzes and 2.1M practice questions across 500+ global exams.

⏱️ ~6 min read

AP Psychology – Statistics in Psychology (Descriptive, Inferential, p?value)

AP Psychology: Statistics in Psychology (Descriptive, Inferential, p-value) – Exam-Ready Study Guide

What This Is

Statistics in psychology help researchers describe data (descriptive stats) and draw conclusions about populations from samples (inferential stats). This topic is high-yield on the AP exam—expect multiple-choice questions on definitions, FRQs asking you to interpret graphs or p-values, and experimental design questions. Real-world example: In the 1950s, psychologist Solomon Asch used statistics to show that 75% of participants conformed to an obviously wrong group answer at least once in his famous line-judgment experiments. Without stats, he couldn’t have proven conformity was a real effect, not just random guessing.


Key Terms & Concepts

  • Descriptive Statistics: Numbers that summarize a dataset (e.g., mean, median, mode, range, standard deviation).
  • Example: If a class’s test scores are [80, 85, 90, 95, 100], the mean (average) is 90.

  • Mean (? or x?): The average of a dataset. Formula: Mean = (Sum of all scores) / (Number of scores)

  • Example: For [3, 5, 7], mean = (3+5+7)/3 = 5.

  • Median: The middle number when data is ordered. If even # of scores, average the two middle numbers.

  • Example: For [3, 5, 7], median = 5. For [3, 5, 7, 9], median = (5+7)/2 = 6.

  • Mode: The most frequent score in a dataset.

  • Example: In [2, 2, 3, 4], mode = 2.

  • Range: The difference between the highest and lowest scores.

  • Example: For [10, 20, 30], range = 30 – 10 = 20.

  • Standard Deviation (? or s): Measures how spread out scores are from the mean. Low SD = data clustered close to mean; high SD = data spread out.

  • Formula (for sample): s = ?[?(x – x?)² / (n – 1)]

    • x = each score, x? = mean, n = number of scores.
  • Normal Distribution (Bell Curve): A symmetric, bell-shaped curve where ~68% of data falls within 1 SD of the mean, ~95% within 2 SDs, ~99.7% within 3 SDs.

  • Example: IQ scores are normally distributed (mean = 100, SD = 15).

  • Inferential Statistics: Techniques that let researchers generalize from a sample to a population (e.g., t-tests, p-values).

  • Example: If a drug trial shows 60% of 100 patients improved, inferential stats help determine if this effect is real or due to chance.

  • Statistical Significance: A result is unlikely due to chance (usually p < 0.05).

  • Example: If p = 0.03, there’s a 3% chance the results are random.

  • p-value: The probability that results occurred by chance. Lower p = stronger evidence against the null hypothesis.

  • AP Exam Tip: p < 0.05 = statistically significant (reject null hypothesis).

  • Null Hypothesis (H?): The default assumption that there’s no effect (e.g., "The drug has no impact on memory").

  • Example: In a study on caffeine and alertness, H? = "Caffeine has no effect on alertness."

  • Alternative Hypothesis (H?): The researcher’s prediction (e.g., "The drug improves memory").

  • Example: H? = "Caffeine increases alertness."

  • Type I Error: False positive—rejecting H? when it’s true (e.g., saying a drug works when it doesn’t).

  • Example: Convicting an innocent person (H? = "not guilty").

  • Type II Error: False negative—failing to reject H? when it’s false (e.g., missing a real effect).

  • Example: Letting a guilty person go free.

Step-by-Step: How to Analyze a Psychology Study’s Stats

  1. Identify the variables:
  2. Independent variable (IV): What’s manipulated (e.g., caffeine vs. placebo).
  3. Dependent variable (DV): What’s measured (e.g., test scores).

  4. Check descriptive stats:

  5. Look at mean, median, mode to see central tendency.
  6. Check standard deviation to see variability.

  7. Determine if results are significant:

  8. Find the p-value (usually given in the study).
  9. If p < 0.05, reject H? (results are significant).
  10. If p > 0.05, fail to reject H? (results may be due to chance).

  11. Interpret the effect size (if given):

  12. Cohen’s d: Measures strength of effect (small = 0.2, medium = 0.5, large = 0.8).
  13. Example: A study with p = 0.01 and d = 0.7 means strong, significant results.

  14. Watch for errors:

  15. Type I Error? Could the results be a fluke?
  16. Type II Error? Was the sample size too small to detect a real effect?

Common Mistakes

  • Mistake: Confusing descriptive and inferential stats.
  • Correction: Descriptive stats summarize data (e.g., mean, SD); inferential stats make predictions (e.g., p-values, t-tests).

  • Mistake: Thinking p = 0.05 means a 95% chance the hypothesis is true.

  • Correction: p = 0.05 means 5% chance the results are due to random variation—not proof of the hypothesis!

  • Mistake: Ignoring effect size and only looking at p-values.

  • Correction: A tiny effect (e.g., d = 0.1) can be "significant" (p < 0.05) if the sample is huge—but it may not matter in real life.

  • Mistake: Assuming correlation = causation because of a low p-value.

  • Correction: Even if p < 0.01, a study can’t prove causation without experimental control (e.g., random assignment).

  • Mistake: Misinterpreting standard deviation as "average distance from the mean."

  • Correction: SD is the square root of the average squared distance—not a simple average.

AP Exam Insights

What’s Frequently Tested? - Multiple-choice: Definitions (e.g., "What is standard deviation?"), interpreting p-values, identifying Type I/II errors. - FRQs: Often ask you to interpret a graph (e.g., "Is the difference between Group A and Group B statistically significant?") or design an experiment (e.g., "How would you test if caffeine improves memory?"). - Tricky Distinction: Statistical significance-practical significance. A study can have p < 0.001 but a tiny effect size (e.g., a drug that "works" but only improves scores by 1%).

Common Traps: - Confusing mean, median, and mode (e.g., picking median when mean is asked). - Assuming p < 0.05 means "the hypothesis is proven" (it just means the results aren’t random). - Forgetting that correlation-causation (even with a low p-value).

FRQ Tip: If asked to evaluate a study’s statistical validity, mention:
1. Sample size (small = less reliable).
2. p-value (is it < 0.05?).
3. Effect size (is the effect meaningful?).
4. Potential confounds (e.g., placebo effect, experimenter bias).


Quick Check Questions

1. Multiple Choice

A researcher finds that students who study with music score 5 points higher on a test than those who study in silence (p = 0.04). What does this p-value mean? A) There is a 4% chance the results are due to the music. B) There is a 4% chance the results are due to random variation. C) The results are 96% likely to be true. D) The effect size is 0.04.

Answer: B Explanation: A p-value of 0.04 means there’s a 4% probability the results occurred by chance (not that the hypothesis is 96% true).*


2. Short FRQ

A psychologist tests whether a new therapy reduces anxiety. After 6 weeks, the therapy group’s mean anxiety score is 45 (SD = 5), and the control group’s mean is 50 (SD = 6). The p-value is 0.02. a) Is the difference statistically significant? Explain. b) What is one potential Type I error in this study?

Answer: a) Yes, the difference is statistically significant because p = 0.02 < 0.05, meaning there’s only a 2% chance the results are due to random variation. b) A Type I error would be concluding the therapy works when it actually has no effect (false positive).


Last-Minute Cram Sheet

  1. Descriptive stats = summarize data (mean, median, mode, SD, range).
  2. Inferential stats = generalize from sample to population (p-values, t-tests).
  3. p < 0.05 = statistically significant (reject null hypothesis).
  4. Standard deviation (SD) = how spread out scores are from the mean.
  5. Normal distribution: 68% within 1 SD, 95% within 2 SDs, 99.7% within 3 SDs.
  6. Type I error = false positive (rejecting H? when true).
  7. Type II error = false negative (failing to reject H? when false).
  8. Effect size (Cohen’s d) = strength of effect (small = 0.2, medium = 0.5, large = 0.8).
  9. p-value-effect size (a tiny effect can be "significant" with a huge sample).
  10. Correlation-causation (even with p < 0.01).