Professional Documents
Culture Documents
Statistics Presentation
Statistics Presentation
By: Tausif Khan, Sayak Maity, Jakob Boeye, and Max Shen
Topic ● Testing soil pH in several
locations in Howard
County
● Objective: to find out if
the average pHs of
different regions of HoCo
are the same
Glenelg Countryside (GC)
Data
Soil pH
Data
Soil pH in Howard County (overall)
6-inch data
χ2 test of homogeneity
No clear, significant
χ2 and t: 3 inch vs. 6 inch difference between 3
inch depth sample pH
and 6 inch depth
sample pH. However,
assumptions for χ2 3 inch 6 inch
test and t-test are both
5.5 2 1
violated. Data for each
individual region 6 5 7
(boxplots, t-tests) says
otherwise. 6.5 6 5
t test 7 6 5
7.5 4 4
mean
difference 0.0242 8 1 2
df 5
df 46
p 0.94659252
p 0.9084
Four Factor Anova
Test with
replication
Parameter: μ equals the average soil
Parameter and pH of all the regions
Hypothesis:
Four Factor ANOVA
H0: The sample mean of all groups is
(ANalysis Of the same
VAriance) Test with
replication HA: The sample mean of all groups is
different
Assumptions for ● Dependent variable is continuous-
fulfilled
ANOVA ● Independent variable consists of
at least two different groups-
fulfilled
● Samples are independent- fulfilled
● Distribution of residuals is normal-
see graph (fulfilled)
● No significant outliers-see graph
(fulfilled)
● Variances are homogenous
(homoscedastic)-see graph
(fulfilled)
First analysis - Sample
ANOVA Test H0: The means of the observations
grouped by location are the same
HA: The means of the observations
grouped by location are different
● The f-statistic is 32.671
Conclusion ● P(f*>32.671) = 7.64E-11
- first analysis ●
●
Let a=0.01
Since 7.64E-11 < 0.01, we will reject
the null hypothesis
● If we assume the null hypothesis that
the average pH for all the groups is the
same is true, than the results obtained
in this study will be observed 7.64E-11,
or close to 0% of the time. Since this is
improbable it is more likely that the
alternative hypothesis is true and that
the average pH of the groups is not the
same.
Second analysis - Columns
ANOVA Test H0: The means of the observations
grouped by depth are the same
HA: The means of the observations
grouped by depth are different
● The f-statistic is 0.403
Conclusion ● P(f*>0.403) = 0.8418
- second analysis ●
●
Let a=0.01
Since 0.8418 > 0.01, we will fail to
reject the null hypothesis
● If we assume the null hypothesis that
the average pH for all the groups is the
same is true, than the results obtained
in this study will be observed about
84.18% of the time. Since this is very
probable, we are unable to reject the
null hypothesis that the means of the
observations grouped by depth are
statistically the same
Since the samples are not in the same
http://www.cropnutrition.com/efu-soil-pH
https://www.nrcs.usda.gov/Internet/FSE_MANUSCRIPTS
/maryland/MD027/0/MDHoward5_08.pdf