- How will this correction affect the mean and the standard deviation of the dataset?
Answer: The mean will increase, and the standard deviation will decrease.
The original mean is (6+8+10+13+18)/5 = 55/5 = 11. The new mean is (9+8+10+13+18)/5 = 58/5 = 11.6. So, the mean increases. The standard deviation measures the average distance of scores from the mean. The original scores are more spread out. By changing the 6 to a 9, that data point moves closer to the mean, making the overall set of scores less dispersed. Therefore, the standard deviation will decrease.
- Which of the following best describes the most likely shape of the distribution of completion times?
Answer: Negatively skewed, because a small number of very short times will pull the mean to the left of the median.
When analyzing distribution shapes, you need to consider where most data points cluster and how outliers affect the overall pattern. The key is understanding how extreme values on one side create "skew" by pulling the distribution's tail in that direction.
In this scenario, most participants complete the puzzle in reasonable time, creating a main cluster of scores. However, some participants give up very early, producing unusually short completion times. These early quitters create extreme low values that form a tail extending toward the left (shorter times), while the main body of data sits toward the right (longer, more typical completion times). This pattern defines negative skew - where the tail points left and extreme low values pull the mean below the median.
Choice A incorrectly describes positive skew, where the tail extends right with high-end outliers. Choice B suggests bimodal distribution, but the passage indicates most people solve it normally with only a small fraction giving up early - this creates one main peak with a tail, not two distinct peaks. Choice C assumes symmetry, but the presence of early quitters on only one end of the distribution eliminates any balance between early and late finishers.
Choice D correctly identifies negative skew. The small number of very short times (from people giving up) creates the left tail that pulls the mean toward lower values, positioning it left of the median.
Remember: skew direction follows the tail. Negative skew = left tail = mean pulled below median. When you see "small fraction" creating extreme values on one side, expect skewed distribution in that direction.
- What is the overall mean exam score for the entire 50-student class?
Answer: 82.4
This problem requires calculating a weighted mean, as the two groups have different sizes. A simple average of 88 and 78 (which is 83) would be incorrect. The calculation is: [(Group 1 size × Group 1 mean) + (Group 2 size × Group 2 mean)] / (Total size). This is [ (20 × 88) + (30 × 78) ] / (20 + 30) = [ 1760 + 2340 ] / 50 = 4100 / 50 = 82.4.
- If the researcher were to incorrectly include this outlier in the final data analysis, what would be the most likely consequence for the measures of central tendency?
Answer: The mean would increase significantly, while the median would remain relatively stable.
The mean is highly sensitive to extreme outliers because its calculation includes the value of every data point. A very large outlier like 5,500 ms will pull the mean upwards significantly. The median, which is the middle value of a sorted dataset, is resistant to outliers. Since only one data point out of 21 is extreme, the middle value will shift only slightly, if at all. Therefore, the mean will increase substantially while the median remains relatively stable.
- Due to the statistical principle of regression to the mean, what is the most probable outcome for the average BDI score of this specific group when measured the second time?
Answer: It will likely be lower and closer to the overall average BDI score of the entire client population.
Regression to the mean is a fundamental statistical principle you'll encounter frequently in psychology research. When you select individuals based on extreme scores (like the top 5% on any measure), their follow-up scores will typically move closer to the population average, even without any intervention.
Here's why this happens: extreme scores often result from a combination of true ability/condition plus measurement error or temporary factors. On retesting, while the underlying condition may remain similar, the random fluctuations that contributed to the extremely high initial scores are unlikely to repeat in exactly the same way.
In this scenario, the severely depressed clients were selected precisely because they had the highest BDI scores. When retested six weeks later, their average score will most likely decrease toward the population mean, making answer A correct.
Answer B incorrectly assumes that untreated depression will automatically worsen over six weeks, ignoring the statistical principle at play. Answer C wrongly suggests that without intervention, scores remain identical – this misses how regression to the mean operates independently of treatment effects. Answer D partially recognizes that individual scores will vary, but incorrectly assumes these changes will cancel out perfectly, leaving the group mean unchanged.
The key insight is that regression to the mean is a measurement phenomenon, not a clinical one. Even if the clients' actual depression levels haven't changed, their group average score will likely appear lower simply due to this statistical principle.
Study tip: Whenever you see extreme groups selected for research, expect regression to the mean on follow-up measurements, regardless of interventions.
- Which participant's score is the most atypical or extreme when compared to the performance of the group?
Answer: Carla
To determine which score is most atypical, we can calculate how many standard deviations each score is from the mean (a z-score). The formula is z = (score - mean) / SD.
Anjali: (280 - 250) / 20 = 30 / 20 = 1.5 SD from the mean.
Ben: (240 - 250) / 20 = -10 / 20 = -0.5 SD from the mean.
Carla: (300 - 250) / 20 = 50 / 20 = 2.5 SD from the mean.
David: (255 - 250) / 20 = 5 / 20 = 0.25 SD from the mean.
Carla's score of 300 is 2.5 standard deviations from the mean, which is the largest deviation among the four participants, making her score the most atypical.
- The graph below shows the mean creativity scores for a control group and a group that received a 'mindfulness training' intervention. The error bars represent the 95% confidence interval for the mean. What is the most methodologically sound conclusion a researcher can draw from this graph?
Answer: The study is inconclusive, as the significant overlap in the confidence intervals suggests the difference may be due to chance.
In psychological research, when the 95% confidence intervals of the means for two independent groups overlap substantially, it is a strong indication that the difference between the group means may not be statistically significant. While the training group's mean is higher, the overlap suggests that the true population means could be identical. Therefore, the most cautious and sound conclusion is that the result is inconclusive pending a formal statistical test. Simply observing a higher sample mean is insufficient proof of effectiveness. Confidence intervals reflect uncertainty about the mean, not necessarily the homogeneity of the sample itself (that's SD).
- A school psychologist compares two different anti-bullying programs, Program A and Program B, in two separate but demographically similar schools. After six months, she measures the number of reported bullying incidents. The results are summarized in the bar chart below, where the error bars represent the standard deviation of the monthly incident counts. Based on the graph shown:
Answer: The average number of incidents was similar for both programs, but the number was more variable for Program B.
The graph shows that the mean (height of the bars) number of incidents is identical for both programs (Mean = 10). However, the standard deviation (represented by the error bars) is much larger for Program B than for Program A. This indicates that while the average outcome was the same, the month-to-month number of incidents was more variable and less consistent for Program B. Stating that Program A is 'superior' is an interpretation that goes beyond the data, which only speaks to consistency, not necessarily quality. Equal effectiveness is incorrect because it ignores the difference in variability.
- What is the most nuanced interpretation of these results?
Answer: The new curriculum was likely beneficial for some students but may have been detrimental or less helpful for others.
When analyzing experimental results in psychology, you need to look beyond just the average (mean) to understand the full picture. The standard deviation tells you how spread out the scores are—low standard deviation means most students scored similarly, while high standard deviation means scores were widely varied.
The correct answer is D because the data shows a complex pattern. While the average increased by 15 points (suggesting overall improvement), the standard deviation jumped dramatically from 8 to 20 points. This means the scores became much more spread out. The most likely explanation is that the new curriculum worked well for some students (who scored much higher) but didn't work as well for others (who may have scored lower or showed little improvement), creating this wider distribution of outcomes.
Answer A is wrong because it ignores the increased variability—a truly successful intervention would ideally help most students without creating such dramatic differences. Answer B incorrectly dismisses the substantial changes as random chance; a 15-point mean increase and more than doubled standard deviation are too large to ignore. Answer C misunderstands what standard deviation measures—it doesn't indicate test reliability, but rather the spread of actual performance.
When you encounter research results on psychology exams, always examine both measures of central tendency (like mean) and measures of variability (like standard deviation) together. Real-world interventions often have differential effects across populations, and recognizing this nuanced pattern—rather than looking for simple "success" or "failure"—demonstrates sophisticated statistical thinking.
- Which of the following statements provides the most accurate statistical comparison of the two groups?
Answer: Group 2 has a higher mean vocabulary, but Group 1 has greater variability.
First, calculate the mean for each group. Mean(1) = (25+50+52+55+58+60)/6 = 300/6 = 50. Mean(2) = (40+45+50+55+60+65)/6 = 315/6 = 52.5. So, Group 2 has a higher mean. Next, assess variability. A simple way is to look at the range. Range(1) = 60-25 = 35. Range(2) = 65-40 = 25. Group 1 is more spread out due to the low score of 25. Therefore, Group 1 has greater variability (a higher standard deviation) despite having a lower mean. So, Group 2 has a higher mean, but Group 1 has greater variability.
- The box plot below summarizes the final exam scores for a large psychology class. Based on the plot, which statement is unequivocally true?
Answer: The range of the scores for the middle 50% of students is 22 points.
A box plot displays the five-number summary. The 'box' represents the interquartile range (IQR), which contains the middle 50% of the data. The bottom of the box is the first quartile (Q1) and the top is the third quartile (Q3). Here, Q1=68 and Q3=90. The range of this middle 50% is Q3 - Q1 = 90 - 68 = 22 points. The line inside the box is the median (82), not the mean. The median (82) is not centered in the box [68, 90] and the lower whisker is longer than the upper whisker, suggesting a negative skew. Approximately 75% of students scored above 68 (Q1), not below.
- A researcher plots the data from a very difficult exam where most students scored poorly, but a few students performed exceptionally well. The resulting distribution is shown in the histogram below. Which statement about the measures of central tendency for this distribution is most likely to be true?
Answer: The mean is greater than the median, which is greater than the mode.
The histogram described shows a positive skew (or is skewed to the right). The 'tail' of the distribution is on the right side, created by the few exceptionally high scores. In a positively skewed distribution, the mode (the most frequent score) is the peak of the distribution, which will be the lowest value among the three measures. The median is pulled slightly toward the tail, and the mean is pulled most strongly toward the tail by the high-scoring outliers. Thus, the relationship is Mean > Median > Mode.
- After adding 2 points to every participant's score, what will the new mean and standard deviation be?
Answer: Mean = 32, Standard Deviation = 8
When a constant value is added to every score in a dataset, the mean increases by that same constant amount. So, the new mean will be 30 + 2 = 32. However, adding a constant does not change the spread or dispersion of the scores relative to each other. Every score shifts up by the same amount, so the distances between the scores and the distance of scores from the mean remain the same. Therefore, the standard deviation is unchanged. The new SD will still be 8.
- A researcher conducts an experiment on memory, examining how word length (Short vs. Long) and presentation style (Auditory vs. Visual) affect recall. The mean number of correctly recalled words for each condition is shown in the table. The data pattern in the table suggests:
Answer: an interaction between word length and presentation style.
An interaction effect occurs when the effect of one independent variable (e.g., presentation style) depends on the level of another independent variable (e.g., word length). Here, for short words, the visual presentation (Mean=15) is much better than auditory (Mean=10). However, for long words, the visual presentation (Mean=8) is only slightly better than auditory (Mean=7). The advantage of visual presentation is much larger for short words than for long words. This differential effect signifies an interaction.
- A clinical researcher compares three therapeutic approaches for social anxiety. The summary statistics for the primary outcome measure (a social anxiety scale) are presented in the table. Based on the table provided, which of the following conclusions is most justified?
Answer: Approach A and Approach C produced results with a similar degree of patient-to-patient variability.
The standard deviation (SD) is a measure of variability or dispersion of scores. A smaller SD means scores are more clustered around the mean (more consistent), while a larger SD means scores are more spread out. Approach A has SD = 10.1 and Approach C has SD = 10.3. These values are very similar, indicating a comparable level of variability in patient outcomes for these two groups. Approach B has the largest SD (15.2), indicating the most variability, not the most similar outcomes. We cannot determine effectiveness from this table without knowing whether high or low scores are better, and without statistical tests. The sample sizes (N) are identical.
- Given this bimodal distribution, what is the most significant limitation of reporting only the mean reduction of 50%?
Answer: The mean is misleading because it represents an outcome that is typical for very few, if any, of the actual patients.
When you encounter questions about statistical distributions and measures of central tendency, always consider whether the mean actually represents what's typical in your data set.
In this bimodal distribution, patients cluster at two extremes: about half experience 90% improvement while half experience only 10% improvement. The calculated mean of 50% is mathematically correct, but it represents an outcome that virtually no actual patients experienced. This makes the mean deeply misleading as a summary statistic because it suggests a "typical" result that doesn't reflect anyone's real experience.
Choice A correctly identifies this core problem - the mean fails to capture what's actually happening with individual patients. When distributions are bimodal or highly skewed, the mean can represent a value that few or no participants actually experienced.
Choice B is wrong because the mean calculation itself is accurate; the issue isn't computational error but rather the appropriateness of using the mean for this type of distribution.
Choice C incorrectly assumes bimodal distributions indicate insufficient sample size. Bimodal distributions can occur with adequate samples and often reveal meaningful subgroups or different response patterns.
Choice D confuses the roles of descriptive statistics. The standard deviation measures variability, not central tendency, so it cannot replace the mean as a measure of central tendency. Additionally, reporting standard deviation alone wouldn't solve the fundamental problem.
Study tip: When you see bimodal or skewed distributions on psychology exams, question whether the mean tells the whole story. Consider what additional statistics (like medians, modes, or subgroup analyses) might better describe the data pattern.
- A student reading the paper wants to understand how much the performance scores typically varied from person to person within the sample. Which value should they focus on?
Answer: The standard deviation (16), as it measures the average dispersion of individual scores around the sample mean.
This question tests the distinction between standard deviation (SD) and standard error of the mean (SEM). The SD describes the variability or spread of data points within a single sample. The SEM describes the theoretical variability of sample means if the study were repeated many times; it is a measure of the precision of the sample mean as an estimate of the population mean. To understand how scores varied from person to person in the actual study, the standard deviation is the correct statistic.
- The scatterplot below displays the relationship between hours of sleep the night before an exam and the score on that exam for a group of college students. A line of best fit is included. The point labeled 'P' represents a particular student. This student's outcome is best characterized as:
Answer: performing better on the exam than would be predicted based on their hours of sleep.
The line of best fit represents the predicted exam score for any given amount of sleep. Points above the line represent individuals who performed better than predicted by the trend, while points below the line represent those who performed worse than predicted. Since point P is located well above the line of best fit, that student achieved a much higher exam score than one would expect given the amount of sleep they got.
- Based on the properties of the normal curve, what percentage of individuals would be expected to score between 84 and 132?
Answer: 81.5%
This requires using the empirical rule (68-95-99.7 rule). A score of 84 is one standard deviation below the mean (100 - 16 = 84). A score of 132 is two standard deviations above the mean (100 + 2*16 = 132). The area between -1 SD and the mean is approximately 34% (half of 68%). The area between the mean and +2 SD is approximately 47.5% (half of 95%). The total percentage is the sum of these two areas: 34% + 47.5% = 81.5%.
- A researcher develops two new scales to measure conscientiousness, Scale X and Scale Y. The histograms below show the distribution of scores from a large sample of adults for each scale. Both distributions have a mean score of 50. Based on the histograms provided:
Answer: Scale Y has a larger standard deviation than Scale X.
Standard deviation is a measure of the average dispersion or spread of scores around the mean. In the histograms, the scores for Scale X are tightly clustered around the mean of 50, indicating low variability. The scores for Scale Y are much more spread out across the possible values, indicating high variability. Therefore, Scale Y has a larger standard deviation. The distributions are symmetric, so the median equals the mean for both. The range for Y is clearly larger than for X.
