GEDU 6170 Research LiteracyQuantitative and Qualitative Research
Saad Chahine, PhD May 6, 2014
Quantitative Research
“technical literacy”- Focus on the specific analytical procedures and how & when to use them
“intuitive literacy”- Focus on a general understanding of the kinds of intuitions needed to understand the statistics
(Shank & Brown, 2007, p. 38)
Statistical Worldview
• Newton example… • By conducting several experiments, developed
an underlying model that can explain gravity • The model can then be used to predict any
falling object • Very deterministic – educational research likes
to be deterministic…but it is difficult to find such absolutes – life is much more about probability
Data is Pervasive
• All observations in life can be thought of a data
• Each observation is a datum • When combined these become distributions • Based on the kinds of data collected, different
distributions can form
Distributions
• Constant Distribution (AKA Uniform Distributions) • “Blob” Distribution (AKA Correlation r=0) • Normal Distribution (AKA Bell Curve) • Systematic Distributions (e.g. t distribution) • Skewed Distributions* • Many more…
(Shank & Brown, 2007)
Uniform Distribution
http://en.wikipedia.org/wiki/File:Uniform_Distribution_PDF_SVG.svg
Correlation
http://en.wikipedia.org/wiki/File:Correlation_examples2.svg
Normal Distribution
http://en.wikipedia.org/wiki/File:Standard_deviation_diagram.svg
t distribution
http://en.wikipedia.org/wiki/File:T_distribution_1df_enhanced.svg
Skewed Distribution
http://en.wikipedia.org/wiki/File:Negative_and_positive_skew_diagrams_(English).svg
Levels of Measurement
• Categorical Data– Non-ordered data – Often represents different categories: sex, eye
colour, SES, and group type (experimental or control)
– An average would be meaningless– More meaningful to talk about different categories
Levels of Measurement
• Ordinal Data– Distance between data points will vary – Examples: placement in a race, survey response,
teacher grades – Averages are not meaningful; middle number
(median) is most representative of data set
Levels of Measurement
• Interval Data– Very similar to ordinal data, however, distances
between points are equal – E.g., temperature and well designed rating scales – Important: ‘0’ is not meaningful – Averages (mean) is meaningful way to describe a
data set
Levels of Measurement
• Ratio Data– Same as interval except the “0” is meaningful – We can say “twice as much” – E.g., Temperature in Kelvin, height, and weight– Average is the most meaning full way to describe
the data set
Central Tendency
• If you want to describe a population or a group of people using one or two numbers you could say:– On average, students in Nova Scotia scored 570 on
an international test of reading (mean)– In Novo Scotia, the most frequent eye colour is
brown (mode) – In a small sub-sample of 10 students, the weekly
time spent on homework was 5 hours (median)
Descriptive vs. Inferential Statistics
• Descriptive statistics describe the sample or population usually by providing values of range, maximum, minimum, central tendency, variance (sum of individual differences from the mean)
• Inferential statistics are often used when you do not have access to the entire population and want to make an inference about this population
Sampling
• Convenience Sample • Purposive Sample • Representative Sample • Random Sample • Can be more complex… e.g., Proportional
Random Sample
(Shank & Brown, 2007, p. 46)
Analytic Procedures
• Correlation• t-test• ANOVA• Chi-Squares • Regression based
(Shank & Brown, 2007, p. 54)
Qualitative Research
• Has varied views and perspectives• More focused on meaning than a quantitative
method • Some basic perspectives that cut across most
qualitative methods
Holistic vs. Experimental
• More focused on examining phenomena in a naturalistic setting
• Less focused on individual components of a complex system
• More focused on interactions with the system as a whole
• Less focused on isolating relationships(Shank & Brown, 2007, p. 60)
Looking for Meaning
• At the most basic level, qualitative research looks for “themes” that describe patterns in a data set
• Researcher can take two different stances: “outsider looking in” vs. “Insider looking
out”• Some researchers can examine self as insider
and outsider in autobiography studies
(Shank & Brown, 2007, p. 62)
Strategies for Data Collection
• Observations • Interviews • Focus groups • Martials analysis • Archival and historical record analysis • Interpretive analysis (e.g. phenomenology) • Participant observations (Shank & Brown, 2007, p.63)
Methods
• Ethnography • Grounded Theory • Case Study • Narrative and Oral Historical Analysis • Critical Theoretical Analysis • Action Research • Qualitative Educational Evaluation (Shank & Brown, 2007, p.65)
Activity
• In groups, review the article you are provided • As a group identify: – Purpose– Methodology– Importance – Relevance to Education