This document summarizes a class on statistical reasoning in AP Psychology. It discusses descriptive statistics like measures of central tendency (mode, mean, median) and how they can be used to describe data distributions. It also discusses measures of variation such as range and standard deviation. Additionally, it covers the concepts of skewed distributions and normal distributions. The document explains how inferential statistics can be used to determine if observed differences between groups are statistically significant or likely due to chance. It emphasizes the importance of representative sampling and cautions against making broad generalizations from a few cases.
This document summarizes a class on statistical reasoning in AP Psychology. It discusses descriptive statistics like measures of central tendency (mode, mean, median) and how they can be used to describe data distributions. It also discusses measures of variation such as range and standard deviation. Additionally, it covers the concepts of skewed distributions and normal distributions. The document explains how inferential statistics can be used to determine if observed differences between groups are statistically significant or likely due to chance. It emphasizes the importance of representative sampling and cautions against making broad generalizations from a few cases.
This document summarizes a class on statistical reasoning in AP Psychology. It discusses descriptive statistics like measures of central tendency (mode, mean, median) and how they can be used to describe data distributions. It also discusses measures of variation such as range and standard deviation. Additionally, it covers the concepts of skewed distributions and normal distributions. The document explains how inferential statistics can be used to determine if observed differences between groups are statistically significant or likely due to chance. It emphasizes the importance of representative sampling and cautions against making broad generalizations from a few cases.
Describing data (descriptive stats) Accurate statistical understanding is important for psychological research Casual estimates often misread reality and misinform Big, round, undocumented numbers warrant caution Teaching statistical reasoning is needed Presentation of statistical information needs more transparency Descriptive statistics are often displayed with a histogram (bar graph) Measures of central tendency: a single score represents a set of scores Mode: Most frequently occurring score(s) in a distribution Mean: Arithmetic average of a distribution, obtained by adding the scores and then dividing by the number of scores; Can be distorted by few atypical scores Median: Middle score in a distribution; half the scores are above it and half are below it. Measures of central tendency
If the mode, mean and median are
identical, the graph will be a normal distribution curve. If they’re different, the graph will be skewed (i.e. lacking symmetry) The mean is the measure that is most susceptible to extreme data points. Positive skew – pulls the mean towards the higher end. Negative skew – pulls the mean towards the lower end. Standard deviation is a more useful measure of variation. A skewed distribution
This graphic representation of the distribution of a village’s incomes illustrates the
three measures of central tendency—mode, median, and mean. Note how just a few high incomes make the mean—the fulcrum point that balances the incomes above and below—deceptively high. Describing variations in data
Measures of variation reveal similarity or diversity in scores.
Range: Difference between the highest and lowest scores in a distribution Standard deviation: a measure of how much scores vary around the mean. Normal curve (normal distribution): Symmetrical, bell-shaped curve that describes the distribution of many types of data Some data, including some psychological data, follow a normal distribution, e.g. height, weight and IQ scores (designed that way). For scientific research, data can still be informative if it’s skewed. However, standard deviation should be cited in such cases, rather than the mean. Income levels are not normally distributed; (they follow a power-law distribution: the Pareto distribution). Different kinds of data The textbook doesn’t mention it but… the types of statistics computed for a set of data depends on which type of data is collected: Nominal data – data that identifies categories, e.g. gender, yes/no answers on surveys, class level in school (e.g. Y10, Y11 and Y12). Ordinal data – data that identifies the order in which a data set falls. Any ranking of items, e.g. your academic ranking at school, is ordinal data. Interval data – includes data that fall within a line that has a zero point (i.e. it terminates at zero). Weight is an example. A weight (i.e. mass) of zero means that a thing doesn’t exist. Height is another example. Ratio data – includes data that fall in a number line (i.e. zero is just another number on the line). For example, a temperature of 0° does not mean there is no temperature (on all temp scales except Kelvin). In psychology, test scores are sometimes considered as ratio data because a score of zero doesn’t usually mean an absence of knowledge…. Reliable differences (inferential stats) Real data are “noisy.” The mean of one group (e.g. breast-fed babies) could differ from the mean of another group (e.g. bottle-fed babies) simply due to random fluctuations in the people sampled. Researchers need to be cautious when they interpret such results. When is an observed difference reliable? Representative samples are better than biased samples. Less-variable observations are more reliable than ones that are more variable. More cases are better than fewer. Generalizations based on a few unrepresentative cases are unreliable. The purpose of inferential statistics is to discover whether our findings can be applied to the larger population from which the sample was collected. Significant differences (inferential stats) Statistical significance indicates the likelihood that a result will happen by chance. When is an observed difference significant? When sample averages are reliable and difference between them is relatively large, the difference has statistical significance. Observed difference is probably not due to chance variation between the samples. In psychological research, proof beyond a reasonable doubt means that the odds of its occurrence by chance are less than 5 percent. The point to remember: Statistical significance does not say anything about the importance of the results. Studying psychology is often extremely hard: “If the human brain were so simple that we could understand it, we would be so simple that we couldn’t.” – quoted by Emerson W. Pugh in The Biology of Human Values Homework
Revise Module 7 for your upcoming quiz
(Optional) Read the College Board’s Curriculum Module: Teaching Statistics