U0, L5: Statistical Analysis Techniques in Psychology

Statistics

Large amount of data can be collected in research studies. Psychologists need to make sense of the data

• Tool to turn data → info

  • Organize & describe in meaningful way

  • Used to make predictions abt pop of interest

Descriptive Statistics

Numerical measures used to summarize & describe characteristics of a dataset

• Help in organizing & presenting large amounts of data in meaningful/concise way

Inferential Statistics

Involves using data from a sample to make inferences/predictions abt larger pop

• Allow researchers to draw conclusions & make generalizations abt pop based on sample data

Measures of Central Tendency

Statistical tools used to describe central/average value of set of data

• Provide a single value that represents “center”/typical value of a distribution

Mean

Measure of central tendency in statistics that represents average value of a set of data

• Calculated by adding up all the values in the data set & then dividing sum by total # of values

Median

Measure of central tendency in statistics that represents the middle value of a data set when the values are arranged in ascending/descending order

• Even # of values: average of the 2 middle values (add together & divide by 2)

Mode

Measure of Central tendency in statistics that represents the most frequency occurring value in data set

• Data set may have 1 mode (unimodal), multiple modes (multimodal), or no mode if all values occur w/ same frequency

Range

Represents difference b/w highest & lowest values in a data set

• To calculate range: subtract lowest value from highest value

• Provides simple way to asses the spread/variability of data by quantifying distance b/w extremes

Normal Curve/Bell Curve

Majority of the data falls near the center (mean) of the distribution, w/ progressively fewer values occurring further away from mean in both directions

Regression to the Mean

When extreme scores tend to get closer to the average when measured again. Happens bc extreme scores = often due to temporary factors, not lasting characteristics

• Important to understand extreme behaviors/performances may not be as consistent over time

Positive Skew

Majority of data clusters in left side, w/ tail extending → right, suggesting presence of outliers/unusually high values

Negative Skew

Majority of data clusters on right side, w/ tail extending → left, suggesting presence of outliers/exceptionally low values

Standard Deviation

Way to measure how spread out/close together #’s are in a group

• Measures how spread out scores are from average (mean)

• Small = most scores are close to average

• Large = scores are more spread out

Percentile Rank

Statistical measure that indicates percentage of scores in a distribution that are equal to/below a particular value

• Commonly used to compare individual’s score w/ those of a larger group/pop

• Tells where you stand compared to others in group

Bimodal Distribution

Type of data distribution in which there are 2 distinct peaks/high points in a histogram/frequency distribution graph

• Indicates that the data has 2 modes/values that occur most frequency

• Suggests data is not evenly distributed & may represent 2 distinct groups/phenomena w/in data set

Statistical Significance

Likelihood that observed results in a research study are not due to chance

• Typically assessed using statistical tests such as t-tests is represented by a p-value

• P-value < 0.05 = statistically significant difference

• P-value > 0.05 = no statistically significant difference

Effect Sizes

Large effect size indicates a substantial difference/relationship, suggesting that the independent variable has a considerable impact on dependent variable

Small effect size indicates a minimal difference/relationship, indicating independent variables has limited effect on dependent variable

• While statistical significance (p values) shows effect exists in study, practical significance (size effects) shows the effect = large enough to be meaningful in real world.

Meta Analysis

Statistical analysis of multiple research studies on same topic to draw overall conclusions

• Combines data from various studies to inc overall sample size & statistical power