Psych HL

Independent Variable (IV)

The variable manipulated or controlled by the researcher; placed on the x-axis.

Dependent Variable (DV)

The variable measured as an outcome of the IV; placed on the y-axis.

Line Graph Purpose

Shows changes or trends over time or across conditions.

Line Graph: Rate of Change

Steep lines indicate rapid change; gradual slopes show slower change.

Bar Chart Purpose

Compares average values across different groups or categories.

Histogram Purpose

Displays frequency distribution of continuous data.

Scatter Plot Purpose

Shows the relationship or correlation between two variables.

Scatter Plot: Strength of Correlation

Tight clusters = strong correlation; wide spread = weak.

Box Plot Purpose

Shows distribution, central tendency, and spread of data.

Box Plot: Interquartile Range (IQR)

Width of the box, showing middle 50% of data.

Box Plot: Range

Length of whiskers showing minimum to maximum values.

Pie Chart Purpose

Displays proportions or percentages of a whole.

Area Graph Purpose

Shows cumulative change over time or stacked category changes.

Area Graph: Segment Trends

How each category increases or decreases over time.

Area Graph: Total Change

Combined area showing overall change.

Line Graph Typical Use

Longitudinal studies showing development or changes over time.

Bar Chart Typical Use

Experiments comparing group performance.

Histogram Typical Use

Descriptive studies showing data distribution (e.g., IQ scores).

Scatter Plot Typical Use

Correlational studies.

Box Plot Typical Use

Studies comparing score distributions across groups.

Pie Chart Typical Use

Survey data showing percentages.

Area Graph Typical Use

Time-series data showing cumulative effects.

P3 Description: Precise Language

Use terms like increases, decreases, stabilizes, peaks, troughs.

P3 Description: Use Data Points

Reference specific values when describing patterns.

P3 Description: Avoid Vague Statements

Be explicit about what the graph shows.

Quantitative data

Numeric measurable data that can be statistically analyzed

4 Levels of quantitative data

Nominal, Ordinal, interval, Ratio

Nominal data

Data based on categories

Ordinal data

Ranked data (strongly agree to strongly disagree)

Interval data

Differences between data, but no true 0 point (the temperature in Fahrenheit)

Ratio data

Has potential score of 0 and you can carry out basic arithmetic on the data (the words remembered or reaction time in seconds)

Qualitative data

Non-numeric descriptive data

Self-report data

where pps tell researchers their experiences in an interview or survey

Behavioral data

obtained when a psychologist observes behavior (eg: reaction time)

Physiological data

biological measures (heart rate)

Archival Data

pre-existing records (medical files)

Anecdotal Data

Someone’s personal experiences or informal observations, not systematically collected, and can be biased

Empirical data

qualitative or quantitative data that is gathered through systematic observations or experimentations that can be independently verified and statistically analyzed

Experimental data

obtained from experiments and allows us to determine causality

Correlation data

quantitative but did not manipulate an IV under controlled conditions with pps randomly allocated

Line Graph Limitations

misleading trends if the axis scaling is improper, difficulty in accurately representing a wide range of data

Bar Chart Limitations

Axis if not scaled correctly can be misleading, can hide variability within a dataset

Histogram limitations

difficult to compare multiple datasets, does not show exact data points only ranges

Scatter Plot limitations

unable to show more than 2 variables at a time, overplotting can occur with a large dataset

Box plot limitations

oversimplifies the data by showing only 5 summary stats (minimum, first quartile, median, IQR, and maximum)

Pie chart limitations

difficulty in accurately comparing slices, especially when they are similar in size

Area graph limitations

difficulty in accurately reading exact values especially for overlapped or stacked data.