t Test for a Mean

Part I: What is t Test?

a. Definition
• The t Test is a statistical test for the mean of a population and is used when the
population is normally or approximately normally distributed and 𝜎𝜎 is unknown.
b. Formula
• The formula for the t test is
𝑋𝑋� − 𝜇𝜇
𝑡𝑡 = 𝑠𝑠
√𝑛𝑛
The degrees of freedom are 𝑑𝑑. 𝑓𝑓. = 𝑛𝑛 − 1.

Part 2: Assumptions for the t Test for a Mean When 𝝈𝝈 Is Unknown (Bluman, 2014)

1. The sample is a random sample.

2. Either 𝑛𝑛 ≥ 30 or the population is normally distributed when 𝑛𝑛 < 30.

Comment: These assumptions are taken from Bluman’s book (See reference). Based on our past
discussion, t will be used if the sample size is small, which is less than 30. However, in the book, one of
the assumptions states that we can also use t if the sample size is large (𝑛𝑛 ≥ 30). The reason we can use
t it is because the 𝜎𝜎 is unknown. Furthermore, I would like to include a part of the text from statlibretext.

Assumptions

The t-test assumes that the observations within each group are normally distributed. If the distribution is
symmetrical, such as a flat or bimodal distribution, the one-sample t-test is not at all sensitive to the non-
normality; you will get accurate estimates of the P-value, even with small sample sizes. A severely
skewed distribution can give you too many false positives unless the sample size is large (above 50 or
so). If your data are severely skewed and you have a small sample size, you should try a data
transformation to make them less skewed. With large sample sizes (simulations I've done suggest 50 is
large enough), the one-sample t-test will give accurate results even with severely skewed data.

Link:

https://stats.libretexts.org/Bookshelves/Applied_Statistics/Biological_Statistics_(McDonald)/04%3A_Tests
_for_One_Measurement_Variable/4.01%3A_One-Sample_t-Test

Part 3: How Do We Perform t Test?

• The process is similar to how we perform z-test. We will only add and adjust certain steps. First,
we will use the t-table now to determine 𝑡𝑡𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐 . Second, to find 𝑡𝑡𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐 , we will use the degrees
𝑋𝑋�−𝜇𝜇
of freedom. Third, the formula for 𝑡𝑡𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐 = 𝑠𝑠 .
√𝑛𝑛

Part 4: Example

Problem: A medical investigation claims that the average number of infections per week at a
hospital in southwestern Pennsylvania is 16.3. A random sample of 10 weeks had a
mean number of 17.7 infections. The sample standard deviation is 1.8. Is there enough
evidence to reject the investigator’s claim at 𝑎𝑎 = 0.05? Assume the variable is normally
distributed.
 STEP 1: 𝐻𝐻0 : 𝜇𝜇 = 16.3 (claim)
𝐻𝐻𝛼𝛼 : 𝜇𝜇 ≠ 16.3
STEP 2: The 𝛼𝛼 = 0.05, 𝑑𝑑. 𝑓𝑓. = 10 − 1 = 𝟗𝟗, and 𝑡𝑡𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐 = ±2.262

a. The confidence level (CL) of an interval estimate of a parameter is the probability that the
interval estimate will contain the parameter, assuming that a large number of samples are
selected and that the estimation process on the same parameter is repeated (Bluman, 2018). The
formula of CL is 𝐶𝐶𝐶𝐶 = 1 − 𝛼𝛼.

If this definition is vague, I will provide an example. Suppose you told your classmate that you can
achieve a score of over 40 in the statistics exam. Then, your friend asks you how sure you are.
You respond, “I am 99% confident.” Do you understand what I'm trying to say? Confidence level
is just a way of saying how sure we are that our estimate is correct. However, in this lesson,
confidence level says how we are so sure that we will not commit a type I error, which is rejecting
the true null hypothesis.

Part 5: How Do We Determine the Critical and Computed Values?

a. To identify the critical values, we will use the z-critical values table for z and the t-table for t.
Moreover, the signs of critical values depend on the tailed-test. We can determine the type of test
by looking at the alternative hypothesis.

Case 1: If right-tailed, the critical value is positive.

Case 2: If left-tailed, the critical value is negative.
Case 3: If it is two-tailed, there are two critical values: one positive and one negative.

b. To know the z-computed, we will follow the formula below.

z for population mean t for population mean z for population proportion

𝑥𝑥̅ − 𝜇𝜇 𝑥𝑥̅ − 𝜇𝜇 𝑝𝑝̂ − 𝑝𝑝
𝑧𝑧 = 𝜎𝜎 𝑡𝑡 = 𝑠𝑠 𝑧𝑧 =
𝑝𝑝𝑝𝑝
√𝑛𝑛 √𝑛𝑛 �
𝑛𝑛

Comment: We will not delve into modifying the formula with a finite population correction factor.

Part 6: How Can We Write Decision and Conclusion?

a. In hypothesis testing, the decisions are to either reject 𝐻𝐻0 or do not reject 𝐻𝐻0 (Fail to reject 𝐻𝐻0 )
b. When writing a conclusion, we will use the format from Triola (2018) outlined below.
 17.7−16.3
STEP 3: 𝑡𝑡𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐 = 1.8 = 2.460
√10
STEP 4: Decision: Reject 𝐻𝐻0
STEP 5: Conclusion: There is sufficient evidence to warrant rejection of the claim.

Part 5: Activity

Instruction: Write your answer on one whole sheet of yellow paper. Follow the procedure based on our
previous meeting. Only the decision and conclusion can be expressed in sentence form. The rest will be
written in symbolic form.

1. The students in a specific strand claim that the average daily sleep time for students in this strand
is not 8 hours. They conducted research that included 25 respondents selected randomly across
10 sections and found an average daily sleep time of 7.08 hours with a sample standard deviation
of 2.53 hours. Assume that the data is normally distributed. At 𝛼𝛼 = 0.01, can it be concluded that
their result differs from their claim?
2. According to a global connectivity intelligence firm from Ookla, the Philippines recorded an
average fixed broadband speed of 93.05 Mbps this January 2024. Random sampling of a large
internet company found that the average fixed broadband speed of 10 households is 89.05 Mbps
with a standard deviation of 23.07. Assume the data is normally distributed. At 𝛼𝛼 = 0.01, can it be
concluded that the average fixed broadband speed has increased?

https://www.pids.gov.ph/details/news/in-the-news/think-tank-says-ph-connectivity-still-lags-
behind-in-asia

3. The average strawberry has approximately 200 seeds. A botanist invented a product that will be
planted with strawberries to see if it can reduce the number of seeds. A random sample of 15
strawberries was selected, and a sample mean of 175.23 seeds with a standard deviation of
23.45 was found. At the 0.05 level of significance, can it be concluded that the mean is less than
200?

Reference:

Bluman, A. (2014). Elementary statistics: A step by step approach 9e. McGraw Hill.