Download Action Research

Survey
yes no Was this document useful for you?
   Thank you for your participation!

* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project

Document related concepts

Foundations of statistics wikipedia , lookup

History of statistics wikipedia , lookup

Misuse of statistics wikipedia , lookup

Time series wikipedia , lookup

Transcript
Chapter 6:
Analyzing Data
Qualitative Data Analysis
Quantitative Data Analysis
Analyzing Data: Qualitative
Qualitative
Data Analysis
Techniques
Inductive
Analysis
Using
Computer
Software
Guidelines
For
Reporting
Analyzing Data: Quantitative
Quantitative
Data Analysis
Techniques
Descriptive
Statistics
Inferential
Statistics
Using
Computer
Software
Guidelines
for
Reporting
Qualitative Data Analysis
Techniques:
 An
inductive process:
(1) Begin w/ specific observations
(2) Note patterns
(3) Formulate tentative hypothesis
(4) General conclusion; theory
 View phenomena from holistic
perspective, factoring in setting,
participants, unique context (Parson &
Brown, 2002)
Qualitative Data Analysis:
Inductive Analysis
 Reduce
large volumes of information.
 Organize the data into important
patterns and themes.
 Being careful not to minimize, distort,
oversimplify, or misinterpret data
(Schwalbach, 2004).

"Systemically organizing & presenting
findings...in ways that facilitate
understanding of data."(Parsons & Brown,
2002)
Qualitative Data Analysis:
Inductive Analysis
 Coding
Scheme: Ways to organize
categories of information; repeated
words or phrases. (Mills, 2003; Schwalbach,
2003)
 Knowing
Your Data: Reading, rereading,
process can be laborious.
 Describe Characteristics of Categories
 Connections between data and research
questions begin to emerge.
Qualitative Data Analysis:
Inductive Analysis
 Reflection:
Describe categories in terms
of their connection to or ability to
understand my research question.
 Conflicting Data: Information that
'conflicts' with patterns.
 Interpret: Examination of events,
behaviors, or observations by category.
 Introspection, Constant Comparisons.
Qualitative Data Analysis:
Inductive Analysis
 Using
Computer Software
 Keyword: 'assist'
 Can help 'sort' information using
electronic 'coding' scheme.
 Useful for large amounts of data.
 Specialized software.
 May need assistance w/ process.
Quantitative Data-analysis
 Descriptive
Statistics: Simple,
mathematical procedures used to
summarize and organize relative large
amounts of numerical data:
(1) Measures of central tendency
(2) Measures of dispersion
(3) Measures of relationship
Measures of Central Tendency
 Mean:
arithmetic average of a set of
scores. May be necessary to drop
'outliers' to get reliable mean.
 Median: specific score in the data set
that separates the entire distribution in
equal halves.
 Mode: most frequently occurring score in
a data set.
Measures of Dispersion
 Measures
of Dispersion: Indicate how
much 'spread' or diversity exist within a
group of scores.
 Range: Distance between highest and
lowest score.
 Standard Deviation: Average distance of
scores away from the mean. 'SD'
impacted by 'extreme' scores.
Measures of Relationship
 Correlation
Coefficients: measures of
direction and degree of relationship
between two variables.
 Strong: 1.00 -- .70 (+ or -)
 Moderate: .70 -- .30 (+ or -)
 Weak: .30 -- .00 (+ or -)
 Direction and Strength of relationship
between two variables.
Visual Representations
 Bar
Graph
 Pie Chart
 Histogram
 Frequency Distribution Table
 Visual way to understand large amounts
of data.
Inferential Statistics

Inferential Statistics: 'Infer' how likely a given
statistical result from a 'sample' applies to an
entire population.


Independent Measures 't' Test: used for 'two
group' (treatment and control). Data are
compared on a common dependent
variable (such as a test score). Mean scores
for two groups are compared to see if
differences are 'statistically' significant. If
difference is 'SS' then there is a 'true'
difference btw groups.
Inferential Statistics
 Repeated
Measures 't' Test: More than
one test score is taken on the same
person in an study.
 'Practical' Significance: subjective
decision of significance determined by
looking at practical factors.
 P-value: indicates probability of chance
occurrences in the study.
Inferential Statistics
 Alpha
level (a): typically set at 0.05 in
educational research studies.
Reasonably certain that only 5% of time
would differences we obtain between
two 'means' be due to chance -- thus
representing no 'real' difference between
the groups.
 If p < a than the difference is statistically
significant. Why?
Inferential Statistics

Is there a Statistically Significant Difference?

Reject the Null Hypothesis: (double negative) Null
Hypothesis says: NO difference between the
groups. So, if we REJECT the Null -- it means THERE
ARE SIGNIFICANT DIFFERENCES BETWEEN THE
GROUPS -- the intervention' made a difference.
(HOORAY!)

Fail to Reject the Null Hypothesis: (triple negative)
= NO significant differences between the groups.
(Project 'failed')
Inferential Statistics
 Analysis
of Variance (ANOVA): variation
of independent 't' test: (or
'True Statistically Significant Differences'
test). Used when there are more than two
'groups' being compared.
 Chi-square Test: Used when looking at
'frequency' counts in data, not scores.
 Ex: Number of boys or girls who...
Using Computer Software
 'StatCrunch'
-- web-based data analysis
software system - Univ. of S. Carolina.
 Fire Up 4.0 Beta! link
 Interactive Java window.
 Feel free to 'experiment' if you are
interested!
 Other software available.
Reporting the Results of
Qualitative Research
 How
do I present the results of my
research most effectively?
 Consider needs of audience.
 Make every effort to be impartial.
 Watch 'value judgments.
 Keep conclusions 'tentative'.
 Provide examples and samples.
Qualitative Research Format:
 Introduction
 Review
of related literature
 Description of innovation/intervention
 Data collection and considerations
 Data analysis and interpretations
 Conclusions
 Reflection and Action Plan
Reporting the Results of
Quantitative Research
 APA
format for reporting numbers.
 Present numerical information in
descending order from largest to least.
 Report total number before small
categories are described.
 Use tables to organize large sets of
numbers or data.
 Use graphs to illustrate numerical data.
References:
1) Mertler, C. A. (2012). Action
Research: Improving Schools and
Empowering Educators, 3rd ed. Los
Angeles, CA: Sage Publishers,