Transcript GLOBALMEDSTATISTICMARCH

Statistics as a Tool
A set of tools for
collecting, organizing,
presenting and
analyzing numerical
facts or observations.
Similar Concepts Across Cultures
Http://www.uptodate.com
Descriptive Statistics

Numerical facts or observations that are
organized describe the frequencies, measures
of central tendency, and degree of dispersion
of variables in a sample of a larger population.
Environmental Studies: Air/Water
Levels of Measurement

Reflects type of information measured and
helps determine what descriptive statistics and
which statistical test can be used.
Four Levels of Measurement
NOIR -- no one is ready
Nominal – lowest level, categories, no rank
Ordinal – second lowest, ranked categories
Interval – next to highest, ranked categories with
known units between rankings
Ratio –
highest level, ranked categories with
known intervals and an absolute zero
Descriptives for nominal and
ordinal data



Frequencies and percentages
Frequencies – absolute number of cases
Percentages – relative number of cases
Descriptives for interval/ratio
(scale) variables

Measures of central tendency
–
–
–
Mean -- sum of all cases divided by number of
cases
Median – case for which half of all other cases are
above and half of all other cases are below.
Mode – most frequently occurring case
Descriptives for scale variables

Measures of dispersion
–
–
–
Range – Value of cases from minimum to maximum
Standard Deviation – number which when added or
taken away from each case adds up to zero.
Variance – Standard deviation squared
Inferential statistics

Procedures used to make inferences from
sample data and generalize findings to the
population
Probability




Statistical significance – the probability that the
difference or the association found in the
sample would be present in the population.
 Three common probabilities used
<.05
<.01
<.001
Sampling bias

The systematic differences between sample in
study and the larger population of interest.

The use of inferential statistics allows us to
calculate the odds that what is found in the
sample is due to sampling bias.
Statistical significance (p-levels)
When p < .05, the degree of difference or
association being tested would only occur by
chance alone five times out of a hundred.
When p < .01, the difference or association being
observed would only occur by chance alone
one time out of a hundred.
When p < .001……
Testing for statistically significant
differences


When you want to see if there is a difference in
outcome by group membership, or by
treatment approach.
SPSS
–
Analyze

Compare means
–
Independent t-test
Planned Parenthood
Earth is in need of another
one/third earth addition: 8 billion
Is there a significant difference in
months of service and type of
outcome?
Group Statistics
MONTHS
Was outcome positive?
yes
no
N
23
14
Mean
11.2174
10.3571
Std. Deviation
5.11643
4.73298
Std. Error
Mean
1.06685
1.26494
Independent Samples Test
Levene's Tes t for
Equality of Variances
F
MONTHS
Equal variances
ass umed
Equal variances
not as sumed
.123
Sig.
.728
t-tes t for Equality of Means
t
df
Sig. (2-tailed)
Mean
Difference
Std. Error
Difference
95% Confidence
Interval of the
Difference
Lower
Upper
.510
35
.613
.8602
1.68725
-2.56506
4.28556
.520
29.309
.607
.8602
1.65476
-2.52258
4.24307
The Need for Pure Water
What cvan be done? Ans:
wells/solar heat
Emerging
Common
Vectors
• Mosquitoes serve
as vectors for
Malaria, Dengue
fever, Yellow
fever, and
Chikungunya
39
• Ticks can serve as
vectors for Lyme
disease, Rickettsia,
and Babesiosis
Statistically significant differences
i.v. nominal and d.v. interval/ratio

Analyze
–
–
–
Univariate (One d.v.; multiple predictors)
Multivariate (Multiple d.v.; multiple predictors)
Repeated measures (time series of dependent
measures; one predictor.
Statistically significant
associations at higher levels of
measurement

Analyze
–
Correlate

Bi-variate
Pearson’s (two interval/ratio variables)
– Kendall’s tau (two ordinal variables)
– Spearman’s rho (two ordinal variables)
–
Test of Pearson Correlation
Coefficient (r)
Correlations
MONTHS
Percent of pos itive cases
for each referral reasons
Pears on Correlation
Sig. (2-tailed)
N
Pears on Correlation
Sig. (2-tailed)
N
MONTHS
1
.
37
.069
.685
37
Percent of
pos itive
cas es for
each referral
reas ons
.069
.685
37
1
.
37
Person-Doctor==Public Health
Group===Population==Mamny
Independent t-test to determine
statistical significance
Independent Samples Test
Levene's Tes t for
Equality of Variances
F
MONTHS
Equal variances
ass umed
Equal variances
not as sumed
.464
Sig.
.509
t-tes t for Equality of Means
t
df
Sig. (2-tailed)
Mean
Difference
Std. Error
Difference
95% Confidence
Interval of the
Difference
Lower
Upper
-13.516
12
.000
-13.0000
.96186
-15.09571
-10.90429
-13.516
10.955
.000
-13.0000
.96186
-15.11809
-10.88191
Differences between groups at
lower levels of measurement

Analyze
–
–
Descriptives…
Crosstabs
–
Identify variable in row and column
– Select statistics
 Two nominal (dichotomized) Chi-Square
 Nominal by ordinal
Kendal’s tau-b
 Nominal by interval
Eta
Difference in LOS by referral
Group Statistics
MONTHS
Reason for referral
mental illnes s
s exual abus e
N
7
7
Mean
5.2857
18.2857
Std. Deviation
2.05866
1.49603
Std. Error
Mean
.77810
.56544
Crosstabs to determine difference
between groups
Reason for referral * Was outcome positive? Crosstabulation
Reas on
for referral
mental illnes s
s exual abus e
physical abuse
neglect
parentreturn
Total
Count
% within Reas on
for referral
Count
% within Reas on
for referral
Count
% within Reas on
for referral
Count
% within Reas on
for referral
Count
% within Reas on
for referral
Count
% within Reas on
for referral
Was outcome
pos itive?
yes
no
6
1
Total
7
85.7%
14.3%
100.0%
5
2
7
71.4%
28.6%
100.0%
5
3
8
62.5%
37.5%
100.0%
3
3
6
50.0%
50.0%
100.0%
4
5
9
44.4%
55.6%
100.0%
23
14
37
62.2%
37.8%
100.0%
Chi-Square tests
Chi-Square Tests
Pears on Chi-Square
Likelihood Ratio
Linear-by-Linear
Ass ociation
N of Valid Cases
Value
3.485 a
3.696
3.334
4
4
Asymp. Sig.
(2-s ided)
.480
.449
1
.068
df
37
a. 9 cells (90.0%) have expected count les s than 5. The
minimum expected count is 2.27.
Which test to use when?




Decision is made by what the question is, the
level of measurement of the variable and the
extent to which assumptions of parametric
statistics are met.
Question: Difference or Association?
Level of measurement: NOIR?
Sample size and distribution (normal?)
Tests comparing difference
between 2 or more groups
Test
Dependent
variable
Paired
Interval/ratio pre
(dependent) t-test and post tests
Unpaired
Interval/ratio
(independent ttest
Independent
variable
Nominal
ANOVA F-test
Interval/ratio
Nominal (>2 grps)
Chi-Square
(Nonparametric)
Nominal
(Dichotomous)
Nominal
Nominal (2 grps)
Tests demonstrating association
between two groups
Test
Dependent var.
Independent var.
Spearman rho
Ordinal
Ordinal
Mann-Whitney U
Non-parametric
Pearson’s r
Ordinal
Nominal
Interval/ratio
Interval/ratio
Tests demonstrating association
between two groups, controlling for
third variable
Test
Dependent
Independent
Logistic
regression
Linear regression
Nominal
Nominal
Interval/ratio
Interval/ratio
Pearson partial r
Interval/ratio
Interval/ratio
Kendall’s partial r
Ordinal
Ordinal
Mission,Manpower,Math,Maintain
How do you handle malnutrition?
Many healthcare task forces
How do you handle AIDS?
Different Theories About HIV/AIDS
Is Health a Right or is Health a
Commodity?
USA Health Care Business Barriers
Some countries will lose
What can be done?