Data Presentation: Histograms, Boxplots and Scatterplots

Exploring Data with Graphs

• Aims of this presentation:
  • Present data clearly
  • Use the Chart Builder
  • Create Graphs: Histograms, Boxplots, Error bar charts, Scatterplots

The Art of Presenting Data

• Graphs should:
  • Display the data.
  • Encourage the reader to contemplate the data being presented, rather than focusing on other aspects of the graph.
  • Avoid distorting information.
  • Present numerous data points with minimal ink.
  • Make large data sets coherent
  • Facilitate the comparison of different data segments.
  • Reveal data.

Chart Builder

• Variables List: Displays all variables available from the data editor.
• Chart Preview: Uses example data to visualize the graph.
• Drop Zones: Areas where variables can be dragged to define axes and data representation.
• Canvas: An example graph appears, allowing users to preview and adjust the layout.
• Gallery: Provides a selection of graph styles for quick selection.

Histograms

• Histograms visually represent:
  • Scores (x-axis)
  • Frequencies (y-axis).
• Histograms help in identifying
  • The shape of the distribution, including skew and kurtosis.
  • The spread or variation in scores
  • Any unusual or outstanding values.

Histograms: Example

• Tested the Disney philosophy that 'Wishing upon a star makes all your dreams come true'.
• Success was measured on 250 people using a composite measure involving salary, quality of life and how close their life matches their aspirations.
  • Measures were taken before intervention and after 5 years
  • Participants were randomly allocated to either, Wish upon a star or work as hard as they can for next 5 years.
• Success was measured using a standardized technique.
  • Score range from 0 to 4
    • 0 = Complete failure
    • 4 = Complete success

Boxplots (Box-Whisker Diagrams)

• Made up of a box and two whiskers.
• The Box indicates:
  • The median value.
  • The upper and lower quartiles.
  • The range containing the middle 50% of the scores.
• The Whiskers point to
  • The range of all scores observed
  • The limits containing the top and bottom 25% of scores.

Error Bar Charts

• These generally show the mean score.
• The error bar projects out from the main bar like a whisker.
• The Error Bar indicates the precision of the measurement in one of three ways:
  • Confidence interval (typically 95%)
  • Standard deviation
  • Standard error of the mean

Bar Charts: One Independent Variable

• Research question: Is there such a thing as a 'chick flick'?
• Participants: 20 men and 20 women.
• Half of each group watched either:
  • A 'chick flick' called The Notebook
  • A control documentary about notebooks.
• Outcome: Physiological arousal served as an measure of enjoyment of the film.

Bar Charts: Two Independent Variables

• Same study as above, but sex is considered as a second independent variable, the Biological Sex of the participant.

Bar Chart: Repeated Measures

• How to cure hiccups?
• All participants experienced each of the 4 interventions in random order
• 15 hiccup sufferers.
• Each intervention:
  • Baseline
  • Tongue-pulling maneuvers
  • Massage of the carotid artery
  • Digital rectal massage.
• Outcome: Number of hiccups in the minute after each procedure.

Bar Chart: Mixed Designs

• Research Question: Is text-messaging bad for your grammar?
• 50 children.
• Split into two groups:
  • Text-messaging allowed.
  • Text-messaging forbidden.
• Each child's grammar was evaluated at two points in time.
  • Baseline/initial assessment.
  • 6 Months later.
• Outcome: Percentage score on a grammar test.

Scatterplots

• Example Topic: Anxiety and Exam Performance
• Participants: 103 students.
• Measures:
  • Time spent revising (hours).
  • Exam performance (%).
  • Exam Anxiety (assessed via the EAQ, score out of 100).
  • Gender