Statistics 1.

T-Test Review
2014. Prepared by Lauren Pincus

With input from Mark Bell

Agriculture Innovation
Program

1
Table of contents
Questions for review .................................................................................................................. 3
Material to Review ..................................................................................................................... 5
1. One-Sample T-Test ......................................................................................................... 5
1.1 Establish Hypotheses ............................................................................................... 5
1.2 Calculate Test Statistic ............................................................................................. 5
1.3 Use This Value to Determine P-Value ...................................................................... 6
1.4 Test your understanding. .......................................................................................... 7
2. Independent (Two Sample) T-Test .................................................................................. 9
2.1 Establish Hypotheses ............................................................................................... 9
2.2 Calculate Test Statistic ............................................................................................. 9
2.3 Use This Value To Determine P-Value ....................................................................10
2.4 Test your understanding. .........................................................................................10
3. Paired T-Test................................................................................................................. 12
3.1 Establish Hypotheses ..............................................................................................12
3.2 Calculate Test Statistic ............................................................................................12
3.3 Use This Value to Determine P-Value .....................................................................13
3.4 Test your understanding. .........................................................................................13
Definition of terms .....................................................................................................................14

2
Questions for review

Situation 1
1. The researcher knows that, nationally, wheat yields are 2,600 kg per ha-1. The
researcher measures wheat yields after side-dressing and finds that yields are
3,085 kg per ha-1 with a standard deviation of 506. If the researcher is interested
in comparing the yields of the side-dressed wheat with national average yields,
then the researcher would want to use a

a) CRD
b) independent t-test
c) paired t-test
d) one-sample t-test
e) RCBD
f) Split plot

2. Which of the following is an appropriate null hypothesis for this study?

a) National yields are greater than the side-dressed wheat yields.

b) The difference between the national average yield and the yields of
side-dressed wheat is zero.
c) Side-dressed wheat yields are greater than the national yields.

3. Write an appropriate alternative hypothesis.

________________________________________________________________
________________________________________________________________
________________________________________________________________

Answer 1
1. d) One-sample t-test
The one-sample t-test compares the mean score of a sample to a known value, usually the
population mean.
2. b) “The difference between the national average yield and the yields of side-dressed wheat is zero.”
3. The difference between the national average yield and the yields of side-dressed wheat is NOT zero.

3
Situation 2
1. A livestock researcher has 20 cows which he randomly divides into two
groups. One group receives a feed supplement and the other does not. Body
weight is measured for both groups after 6 months. Since the researcher is
comparing the means of these two sample groups, the researcher wants to use a

a) CRD
b) independent t-test
c) paired t-test
d) one-sample t-test
e) RCBD
f) Split plot

2. If the researcher stated that the null hypothesis is “body weight of the
supplemented cattle are GREATER than weight from cattle that do not receive a
supplement,” then the researcher would want to use a

one-tailed t-test  Yes  No

two-tailed t-test  Yes  No

Answer 2
1. b) Independent t-test
Use an independent t-test when you want to compare the mean of one sample with the mean of
another sample to see if there is a statistically significant difference between the two. As the name
suggests, you use an independent t-test when your samples are independent
2. One-tailed
A one-tailed test is used for hypotheses which state that the observed mean is either less than or
greater than the true mean (H1: μ > μ0 OR H1: μ < μ0), but not both. If you want to test both greater than
and/or less than (i.e., H1: μ ≠ μ0), then you need to use a two-tailed test.

Situation 3
A researcher believes that tomato variety has an effect on yield response to side-
dressed nitrogen. To account for this, the researcher grows two seedlings each
of 10 tomato varieties. One seedling of each variety gets side-dressed nitrogen
and the other receives no side-dressing. In this study, the observation for each
variety in a sample is matched with the observation for the same variety in the
other sample. To properly compare these two groups (with and without side-
dressing), the researcher would want to use a

a) CRD
b) independent t-test
c) paired t-test
d) one-sample t-test
e) RCBD
f) Split plot

4
Answer 3
3. c) paired t-test
A paired t-test is used to compare two population means where you have two samples in which
observations in one sample can be paired with observations in the other sample.

Material to Review

Test Summary
1. One-Sample T-Test Compare the mean score of a sample to a known
value
2. Independent (two- Compare the mean of one sample with the mean of
sample) T-Test another sample
3. Paired T-Test Determine whether there is a difference between the
average values of paired samples subjected to two
different conditions

1. One-Sample T-Test

The one-sample t-test compares the mean score of a sample to a known value, usually
the population mean (the average for the outcome of some population of interest). The
test is a comparison of the mean of the sample (observed average) and the population
mean (expected average), with an adjustment for the number of cases in the sample
and the standard deviation of the average.

1.1 Establish Hypotheses

The null hypotheses can take one of two forms:
 difference between observed and expected is 0, or
 difference between observed and expected is not 0

1.2 Calculate Test Statistic

Calculation of the test statistic requires four components:

1. The average of the sample (observed average)

2. The number of observations (n).
3. The population average or other known value (expected average)
4. The standard deviation of the sample, or observed, average

With these four pieces of information, we calculate the following test statistic, t:

5
Standard deviation can be calculated using the following equation.

To calculate the t statistic for a sample with 2, 4, 4, 4, 5, 5, 7, 9 as the observations and

4.5 as the known population average.

N (number of observations) = 8
Sample average = (2+4+4+4+5+5+7+9)/8 = 5
Sample standard deviation = (sum(obs-av)2 / # obs)

((2-5)2 + (4-5)2 + (4-5)2 + (4-5)2 + (5-5)2 + etc./8)

((9+1+1+1+0+0+4+16)/8) = 4
 √4
=2

t = (5 - 4.5) / (2 x √ (8 / 7) )
= 0.109

1.3 Use t Value to Determine P-Value

First chose a significance level, often referred to as alpha (α). A critical value of 5%
is usually considered acceptable. This means that we are willing to accept a 5%
chance, or probability, of incorrectly rejecting the null hypothesis. A critical value is
a point on the test distribution that is compared to the test statistic to determine
whether to reject the null hypothesis. Critical values correspond to α, so their values
become fixed when you choose the test's α. If the absolute value of your test statistic
is greater than the critical value, you can declare statistical significance and reject
the null hypothesis.

The p value is the probability of the test statistic being at least as extreme as the one
observed given that the null hypothesis is true. If the p value is equal or smaller than
our chosen significance level, it suggests that the observed data (the values we
collected in our particular sample, is inconsistent with the assumption that the null

6
 hypothesis is true. The null hypothesis must be rejected and the alternative
hypothesis accepted as true. A p value of 0.001 tells us that there is a 0.001%
probability that we mistakenly concluded that our alternative hypothesis is true.
Rejecting the null hypothesis when it is actually true is known as a Type I error.

Use an online calculator such as the one linked here to find the p value
corresponding to our calculated t statistic.

http://www.socscistatistics.com/pvalues/tdistribution.aspx

1.4 Test your understanding.

Question 1: You are interested in knowing how the maize yields in your region compare
to the national average. The national average is 4 ton ha-1. You sample 10 fields in your
region and construct the data set below. Are yields in your region statistically different
than the national average?

Sampled
Yields
3.5
2.7
6.4
3.4
2.9
3.7
3.1
4.3
2.6
3.0

Answer 1:
Sample average = 32.6 / 10 = 3.56
Sample SD = √ ( (3.5 – 3.56)2 + (2.7 – 3.56)2 + (6.4 – 3.56)2 + … + (3.0 – 3.56)2 / 10) =
1.0622
N = 10

t = (3.56 – 4.0) / ( 1.0622 x √(10 / 9) ) = -0.3929

7
Using an online calculator tells us the p value is 0.704179. The result is not significant at
p < 0.05.

Question 2: You are using SPSS to run a one-sample t-test and receive this output.
How would you report the results?

One-Sample Statistics
N Mean Std. Deviation Std. Error Mean
maize_yield 40 3.7225 .73709 .11654

One-Sample Test
Test Value=4
95% Confidence
Interval of the
Sig. (2- Mean Difference
t df tailed) Difference Lower Upper
maize_yield -2.381 39 .022 -.27750 -.5132 -.0418

Answer 2:
Mean regional maize yields (3.73 +/- 0.74) were lower than the national average of 4.0
kg ha-1, a statistically significant difference of 0.26 kg ha-1 (95% CI, 0.04 to 0.51), t(39) =
-2.831, p = .022.

8
 2. Independent (two-sample) T-Test

You use an independent t-test when you want to compare the mean of one sample with
the mean of another sample to see if there is a statistically significant difference
between the two. As the name suggests, you use an independent t-test when your
samples are independent. The two-sample t-test is a hypothesis test for answering
questions about the mean where the data are collected from two random samples of
independent observations, each from an underlying normal distribution.

2.1 Establish Hypotheses

Null hypothesis is that the difference between the two groups is 0. Another way
of stating the null hypothesis is that the difference between the mean of the
treatment group and the mean of the control group is zero.

Alternative hypothesis - the difference between the two sample means is not
zero.

2.2 Calculate Test Statistic

Calculation of the test statistic requires three components:

1. The average of both samples (observed averages). Statistically, we represent

these as

2. The standard deviation (SD) of both averages. Statistically, we represent these

as

3. The number of observations in both populations, represented as

Let’s say an analysis of data comparing side-dressed tomatoes and non side-dressed
tomatoes showed the following:

Side- Non side-

dressed dressed
tomatoes tomatoes
Average 3100 g 2750 g
weight
SD 420 425
N 75 75

9
 2.3 Use This Value To Determine P-Value

Having calculated the t-statistic, follow the same procedure as above to determine
whether the t-statistic reaches the threshold of statistical significance. For independent
samples however, degrees of freedom are calculated by (N1 – 1) + (N2 – 1).

2.4 Test your understanding.

Independent random samples selected from two melon variety populations produced
the results shown below:

Melon Melon
Variety A Variety B
Average 5 kg 7 kg
weight
SD 1.2 kg 0.4 kg
N 42 35

1. What is the null and alternative hypothesis?

2. What is the t statistic?

3. Can the researcher say there is a statistically significant difference between the
weight of these two melon varieties?

Answers:

10
1. Null: There is no difference between the average melon weights.

Alternative: There is a difference between the average melon weights.

2. t statistic = (5 – 7) / √ (1.2 / 42) + (0.4 / 35) ) = -10.15

3. The mean weights between Variety A and Variety B were significantly different (t(75)
= -10.15, p < 0.00001)

11
 3. Paired T-Test

A paired sample t-test is used to determine whether there is a significant difference

between the average values of the same measurement made under two different
conditions. Both measurements are made on each unit in a sample, and the test is
based on the paired differences between these two values.

3.1 Establish Hypotheses

Null hypothesis for the paired sample t-test is H0: d = µ1 - µ2 = 0

where d is the mean value of the difference.

This null hypothesis is tested against one of the following alternative hypotheses,
depending on the question posed:

H1: d > 0
H1: d < 0

The paired sample t-test can only be used when we have matched samples.

3.2 Calculate Test Statistic

We get the following data in our study where a researcher takes 2 transplants from 10
varieties and randomly assigns them to either receive side-dressing or not.

Variety ID Yield of Side- Yield of non Differences Differences^2

dressed side-dressed
tomatoes tomatoes
1 120 94 26 676
2 156 87 69 4761
3 148 118 30 900
4 95 97 -2 4
5 107 110 -3 9
6 98 82 16 256
7 93 105 -12 144
8 114 95 19 361
9 127 134 -7 49
10 134 121 13 169

Means 119.2 104.3 Sum of Sum of

differences: differences^2:
149 7329

12
To work out the differences, the yield of side-dressed tomatoes is subtracted from the
yield of the non side-dressed tomatoes. This value is then squared to produce the
differences2 values.

The test statistic for the paired sample t-test is given by:

t = Σ differences / √((N*Σ differences2 - (Σ differences)2) / (N-1)).

From the table above it can be seen that the Σ differences = 149 and the Σ
differences2 = 7329.

N = 10

t can be computed as:

t = 149 / √(((10 * 7329) - 1492) / 9) = 0.0262

3.3 Use This Value to Determine P-Value

Having calculated the t-statistic, follow the same procedure as above to determine
whether the t-statistic reaches the threshold of statistical significance. For paired
samples however, degrees of freedom are computed as N - 1.

3.4 Test your understanding.

Question 1:
Calculate the t statistic for the following paired samples and determine if there is a
statistically significant difference between the two samples

Sample 1 Sample 2
7 24
5 22
12 23
9 26
5 29
12 22

Question 2:
Is a paired t test needed for the following scenarios? Answer Yes or No.

1. Comparing the average yields of two legume varieties?

2. Comparing weight gain in goats before and after being fed a special diet?
3. Testing whether shoot length and leaf width are related?

13
Answer 1:

The value of t is 7.745967. The value of p is 0.000573. The result is significant at p ≤

0.05.

Answer 2:

1. No
2. Yes
3. No

Definition of terms

Variability is a characteristic of biological material. Hence we need to decide whether

differences between experimental units result from unaccounted variability or real
treatment effects.

An experimental unit refers to the unit of material to which a treatment is applied. It

can be a single leaf, a whole plant, an area of ground containing many plants, a pot, or
in the greenhouse. The term plot is synonymous with experimental unit.

A treatment may be an amount of material or a method that is to be tested and

compared with other treatments in the experiment; e.g., cultivar fertilizer doses, etc.

Variable: A measurable characteristic of an experimental unit is a variable; plant height,

days to flowering, panicle length, or grain yield etc. Individual measurements of a
variable are data; 150 cm plant height, 45 days to 50% flowering, 12 cm panicle length,
or 3240 kg/ha of grain etc.

Data: A set of observations or measurements of a particular variable in an experiment;

105 cm, 95 cm, 120 cm, 75 cm, 100 cm, ...., represent data about the variable (plant
height).

Population: In a statistical sense, a population is a set of measurements or counts of a

single variable on all the units in the specified population.

Sample: A sample is a set of measurements (observations) obtained from part of the

specified population. For example: All 150 plants in a plot form the population. Ten
plants used for recording the plant height from this population form the sample. We
obtain information from the sample.

14
Mean: Mean is the simple arithmetic average (of the sample or population).

Standard Deviation (SD): The measure of dispersion of the data around their mean is
the Standard Deviation. SD tells us how scattered the sample observations are around
the mean.

Variance: Square of the Standard Deviation.

Standard Error (SE): The SD of the population of means is its standard error. SE tells
us how scattered the treatment means (for example) will be.

Degrees of freedom (df): Represents the freedom with which the variability in a data
set could be accounted for. Usually, but not necessarily, df is one less than the number
of observations (n-1).

Distribution of data
Data on a variable may commonly follow one of the following distributions.

Binomial distribution
This type of distribution is expected in data, which describes the proportion of
occurrences in which each occurrence can only be one of two possible
outcomes. For example in data representing percent survival of insects, one can
expect either dead or alive insects.
Poisson Distribution
This is a distribution, which represents occurrences of rare events. For example,
count data such as the number of infested plants, the number of lesions per leaf,
or number of weeds per unit area.

Normal Distribution
These data represent a continuous distribution. Most biological data, when
plotted in a frequency curve, represent a bell shaped and symmetrical curve.
Such data are said to be normally distributed. For example, grain yield, plant
height, etc.

Test of Significance:
Variability is a characteristic feature of nature. Two plants growing side by side are not
alike even under similar conditions. We also know that data on plant growth characters
reveal this variability. For example, from data on five samples plant height such as 115

15
cm, 95 cm, 82 cm, 108 cm, and 72 cm, ....., can we say the growth was good (115 cm
plant height) or the growth was poor (72 cm plant height)? What is the truth?

16