Transcript Online Templates for Basic Statistics

Online Templates for Basic
Statistics:
Rubric Lines 5 & 6 & (4)
Cindy Alonso
David Buncher
AP Research
Common statistics:
Methods, Results, Discussion
•
•
•
•
•
•
•
Number of participants
Mean
Standard deviation
t-tests
ANOVA
Chi Square
Regression and R2
Results
• Presents the findings, evidence,
results, or products
• Tables, data tables, charts, graphs
etc…
• Label all tables and refer to tables
and graphs in the text
• Note the p values, t-values, Fvalues
Discussion/Conclusion
• Interprets the significance of the
p values of the results or findings
• Explores connections to the
original research question
• Include more lit review
• Discuss the implications and
limitations of the research
• Line 5: establish argument from
results
• Line 6: Data analysis from results
• Limitations, future research
Variables
• Independent variable- manipulated
variable
How much water is added to a pea
plant
• Dependent variable- the outcome
How tall the flower grows
• Control- the pea plant without
receiving any water
• Constants- same temperature, light,
etc
• Usually mentioned in “Methods”
Number of participants
• Number of groups
• Number of participants in each
group
• How were the participants selected
• Filtering data: males/females,
AP/non AP, …
• “Methods”
Results: Mean and
standard deviation
• # of participants (N), the “more”
the better (discuss)
• Mean (average) Add up all the
numbers and divided by the # of
responses
• Standard deviation- how spread
out is the data?
• http://www.socscistatistics.com
• Range
• Makes nice looking graphs
and charts for results
t-tests
• Are the two means statistically
(significantly) different?
2 independent means:
• Dominos vs Papa Johns delivery
time
• http://www.socscistatistics.com/t
ests/studentttest/
• 2 dependent means:
• Pretest vs posttest
• http://www.socscistatistics.com/t
ests/ttestdependent/Default.aspx
t-tests
• Null hypothesis: No difference
between the two means
• P level usually < .05: results and
conclusions
• 95% confident of your results
• 1-tailed or 2-tailed outcome
• Bar graphs with p values in results
or conclusions sections
ANOVA
• Are the “greater than two” means
statistically (significantly) different
from each other?
• F statistic, p value
• http://www.danielsoper.com/stat
calc3/calc.aspx?id=43
• Bar graphs in results or
conclusion
Chi Square
• Let’s say you want to know if
there is a difference in the
proportion of men and women
who are left handed and let’s say
in your sample 10% of men and
5% of women were left-handed.
For example, you ask 120 men and 140
women which hand they use and get
this:
Men
Women
LeftRighthanded handed
12
108
7
133
• Interpretation
• Greater differences between
expected and actual data produce
a larger Chi-square value. The
larger the Chi-square value, the
greater the probability that there
really is a significant difference.
• Tables in results
• Discussion of p value in discussion
section
Correlation (Linear
regression)
• Relationship between one
independent variable and one
dependent variable:
• Y = mx +b straight line
• Prediction model
Y = dependent variable
x = independent variable
b = dependent variable when
independent variable = 0 (yintercept)
m= slope !!!
Discussion section
Scatter plot to determine
Correlation
Linear line of best fit y=mx+b
Correlation
• Caution:
• cause and effect
• Obvious relationships: colinear
• R strength of correlation R = 1 is
perfect
• http://www.socscistatistics.com/t
ests/pearson/
• P value
R2
• R-squared (R2) is always between 0
and 100%:
• 0% indicates that the model
explains none of the variability of
the response data around its mean.
• 100% indicates that the model
explains all the variability of the
response data around its mean.
• In general, the higher the Rsquared, the better the model fits
your data.
Good Luck
• David Buncher
• [email protected]
• Cell: 305-527-5000