Week 4 Lesson Videos

Lesson 1 (Chapter 4): Working with SPSS & Lesson 2 (Chapter 5): Descriptive Statistics, Charts, and Frequency Distributions & Lesson 3 (Chapter 5): Running Descriptive Statistics, Charts, and Frequency Distributions in SPSS

variable view

• name: keep it simple (no spaces/symbols)

• type: string or numeric

• decimals: as few as necessary

• label: if extra explanation is needed

• values: for nominal and ordinal categories (ex. 1.00 = vanilla)

• measure:

  • nominal

  • ordinal

  • scale (interval/ratio)

data view

• it looks like Excel, but it’s not

• each cell is a single data point

• cannot add headings

• type in all answers as numbers

• can view data as numbers or value labels

frequency distribution types

• raw data (analyzing data)

• grouped data (present data)

categorical data charts

• pie chart

• bar chart

scale data charts

• histogram

• stem-and-leaf plot

• boxplot

• line chart

• scatterplot

pie charts

display proportions of the total sample that fall into each group

• great for nominal data

• show ordinal data, but does not show rankings well

• more categories → less useful it becomes

bar charts

display each category as separate vertical or horizontal bar

• bar length is determined by frequency for group

• great for both nominal and ordinal data

• no meaningful distance between bar

• confused with histograms, but very DIFFERENT

histograms

display ranges of scores as bars on a continuous scale

• gives a visual of the distribution of the ata (ex. normal, uniform, skewed)

• loses specific frequency data

• great for both interval and ratio data

• meaningful distance displayed between bars

stem-and-leaf plots

display data in rows with stems indicating the first digit(s) and the leaves leaf showing the last digit

• can be confusing to read if unfamiliar with them

• can provide very specific frequency data

• can show some info on the distribution without losing specific frequency data

• can show outliers

boxplot

displaying quartiles using a box to represent the center 50% of data and two whiskers for other 50% of data

• greater for showing the median, interquartile range (height of the box), and outliers

  • max of typical values is 1.5 x interquartile range

• shows how symmetrical or skewed dat is

• can be compared across groups

line chart

displays changes in values over time

• can be very useful for observing trends or can be misleading

scatterplots

displays the relationship between two ranked variables

• plots each observation based on graphical (x, y) points

• best for scale data (interval or ratio), but can be useful for ordinal

• can be used to determine a best fit line for prediction modeling

• only works if you have two variables to compare

frequencies function in SPSS

• click analyze → descriptive statistics → frequencies

  • in window opens, choose the variable(s) that you want to run in the box on the left

  • drag and drop them into the variable(s) box, or select them and click the arrow button to move them to the variable(s) box

  • click the Statistics button

    • in window opens, check all of the descriptive statistics you want to run

    • click continue

  • click the charts button

    • in window opens, select chart type you want (bar chart, pie chart, or histogram)

    • if you want to overlay a normal curve on a histogram, check provided box

    • click continue

  • if you do not want to run a frequency distribution, uncheck the “display frequencies”

  • click “ok”

explore function in SPSS

• click analyze → descriptive statistics → explore

  • in window opens, choose the variable(s) that you want to run in the box on the left

  • drag and rdop them into the Dependent list box, or select them and click the arrow button to move them to the dependent list box

  • do not put any variables into the factor list box unless you want to divide the data for your dependent variables into groups based on a grouping/independent/factor variable. when you first run descriptive statistics, do not do this step. it may come in handy in the group project only

  • you do not need to click the statistics button

  • click the PLOTS button

    • in window open, select stem-and-leaf plot, histogram, or both

    • if you want to run normality plots with tests, check the provided box

    • click continue

  • select the options to run Statistics only, Plots only, or both

  • click ok for results