Research W8: Quantitative Data Analysis Cont'

• Descriptive statistics arrange data visually and help in understanding sample characteristics, summarizing data in manageable portions.
• Descriptive statistics measure:
  • Central tendency
  • Variability (spread)
• Visual representations of data analysis:
  • Scatter plots
  • Histograms
  • Tables
  • Bar graphs
• Inferential statistics find relationships between variables.
• Descriptive statistics summarize data (describing a population).
• Inferential statistics tests for relationships to make predictions (inferences or tests made about a sample population).

Data Analysis

• Data sets are usually in table format using Excel or statistical analysis software.
• Each column is a variable.
• Each row corresponds to a record.
• Responses are entered for each participant.
• Data sets should be complete for analysis.

Data Set

• Data sets ready for analysis are "clean".
• Determining how to clean data will hinge on the level of variable measurement and how variables are classified.

Levels of Measurement for Variables

• The level of measurement determines the type of statistical analysis:
  • Nominal
  • Ordinal
  • Interval
  • Ratio

Nominal Scale

• Classifies objects or events into categories.
• Dichotomous or categorical (e.g., gender, marital status, true/false).

Ordinal Scale

• Shows relative ranking.
• Numbers assigned to categories can be compared.
• Higher ranks have more of an attribute than lower ranks.
• Ex: Ranking by education level, Likert scale.

Interval Scale

• A scale with equal intervals between numbers.
• Zero point is arbitrary (chosen).
• Ex: Temperature, Beck Depression Inventory, test scores.

Ratio Scale

• Ranked on scales with equal intervals and absolute zeros.
• Typically, this is achieved in the physical sciences.
• Ex: Weight, blood pressure, height

Data Set Preparation

• Steps for data set preparation for analysis:
  • Review questionnaire
  • Identify the measurement level for each variable Clean the data file.

Cleaning the Data File

• Steps for cleaning the data file might include:
  • Are there words where numbers should sit?
  • Is all data complete?
  • Is any of the data not in numerical form and need to be put into bins (cluster x organization ej. bin of numbers from 1 to 10, needed for histograms)
• Data shouldn't be altered during the cleaning process.

Data Set - Bins

• Putting data into bins helps to transform the data into a usable format for analysis.
• Researchers will assign a number to each bin so statistical analysis can be performed.

Descriptive Statistical Analysis

• The type of descriptive statistical analysis you can perform is determined by the dataset itself.
• Focus of Descriptive Statistics:
  • Mean
  • Mode
  • Median
  • Range
  • Standard deviation
  • Frequency distribution

Data Analysis Formulas

• Formulas of interest include:
  • AVERAGE: the average for a series (Mean)
  • MIN: lowest score in a series (Range - min)
  • MAX: the highest score in a series (Range - Max)
  • MEDIAN: averages in a series
  • MODE: the number that occurs the most frequently
  • STDEV: standard deviation for a series.
  • FREQUENCY: calculates how often values occur within a range

Data Central Point

• MEDIAN tells the central data point, showing what's typical without being affected by extreme values
• Data should be examined for the overall trend

Deviation

• Low standard deviation = data points near the mean (less variation).
• High standard deviation = data points far from the mean (more variation).

Creating Charts

• Scatter plots, histograms, tables and bar graphs support data interpretation and presentation
  • Format the data that you want to chart
  • Highlight the data range you want to plot
  • Select insert and choose the desired chart type.
• You should also consider the following:
  • Chart type best suited for easing interpretetion
  • The benefit of demographic data put into chart form

Pie Charts

• Pie charts can be used for nominal and some ordinal data
• A circle contains pie-shaped wedges that correspond to percentages for a given category or data.
• Overall Pie charts should add up to 100%, and wedges placed in order, usually with biggest wedge used at 12 o'clock.

Histograms

• Used for itnerval- and ratio-level data.
• Similar to bar graphs, but adjacent values are on a continuum (bars touch one another)
• Data values arrange from lowest to highest and the bars show how much frequency in data
• The horizontal axis shows what the data represents
• The histogram can demonstrate shape of data
• Histograms display data from grouped frequency distributions.

Frequency Polygon

• Used for interval or ratio level data.
• Dots above the score denote frequency.

Distribution

• How the graph looks.
• Is it a normal distribition?

Thematic Analysis

• Looks for recurring topics or themes in reponses, regardless of tone.
• Example is "What common ideas, themes, or concepts do you notice when you read through the comments?"