AP Statistics Topic 4.4: Setting Up a Test for a Population Mean or Population Mean Difference

01 · Start here

The concept in 60 seconds

Setting up a test means deciding what to test and why the method is appropriate. For quantitative measurements from $1$ population with unknown population $\operatorname{SD}$ , use a one-sample $t$ -test for a population mean.

For matched pairs, calculate a difference within each pair. Those differences form $1$ sample, so use a one-sample $t$ -test for the population mean difference.

$1$ mean: $H_0:\mu=\mu_0$ . Choose $H_a:\mu\lt\mu_0$ , $\mu > \mu_0$ or $\mu \ne \mu_0$ to match the investigative question.

Paired mean difference: Define $d$ in a stated order. For no average difference, write $H_0:\mu_d=0$ . Choose the alternative using that same order.

The hypotheses concern population parameters, not the observed sample mean. The sample is the evidence you will use to investigate the hypotheses; it is not the hypothesis itself.

This lesson builds the setup. Topic 4.5 develops the test statistic, $p$ -value and final conclusion.

02 · One situation, deeper understanding

A bottle-filling investigation

Fictional teaching investigation: A team asks whether the mean fill volume of bottles in a particular lot differs from $500 \,\mathrm{mL}$ . It selects an SRS of $25$ bottles without replacement from a lot of $10{,}000$ . The sample mean is $501.2 \,\mathrm{mL}$ and the sample $\operatorname{SD}$ is $7.5 \,\mathrm{mL}$ . Previous process information supports an approximately normal population model. The population $\operatorname{SD}$ is unknown.

The target is $\mu$ , the mean fill volume of all bottles in this lot. Because the question says “differs,” a change in either direction matters:

$H_0:\mu=500\,\mathrm{mL}$

$H_a:\mu\ne500\,\mathrm{mL}$

The method is a one-sample $t$ -test for a population mean. The sample average of $501.2 \,\mathrm{mL}$ does not change the hypotheses to a greater-than alternative. Direction comes from the question, before using the sample result to select it.

Pause and predict: Would a question about underfilling use the same alternative? What would change if the question specifically asked whether the mean is below $500 \,\mathrm{mL}$ ?

Reveal the reasoning

The null stays $H_0:\mu=500$ . For evidence of underfilling, use $H_a:\mu\lt500$ . The sample mean being above $500$ does not justify switching to a different question after seeing the data.

03 · Understand the language

Key ideas and notation

$\mu$ and $\mu_0$

$\mu$ is the unknown population mean. $\mu_0$ is the benchmark mean specified under the null hypothesis.

For the bottles, $\mu_0=500 \,\mathrm{mL}$ . The benchmark comes from the claim, not from $\bar{x}$ .

$\bar{x}$ , $s$ and $n$

The sample mean, sample $\operatorname{SD}$ and sample size. These are observed summaries that later enter the test calculation.

$\bar{x}=501.2 \,\mathrm{mL}$ , $s=7.5 \,\mathrm{mL}$ and $n=25$ . Sample $\operatorname{SD}$ $s$ does not make population $\operatorname{SD}$ $\sigma$ known.

$H_0$ : the null hypothesis

The reference claim used to build the test model. For this AP one-sample setup, write equality to a specified population mean.

$H_0:\mu=500$ . It does not say that every bottle contains exactly $500 \,\mathrm{mL}$ .

$H_a$ : the alternative hypothesis

The population-level pattern the investigation seeks evidence for: less than, greater than or different from the benchmark.

$H_a:\mu\ne500$ is two-sided. $H_a:\mu\lt500$ is one-sided.

$\mu_d$ , $\bar d$ and $s_d$

The population mean, sample mean and sample $\operatorname{SD}$ of paired differences. Define $d$ before choosing the alternative.

For $d=\text{before}-\text{after}$ time, $\mu_d > 0$ corresponds to lower after times on average.

Degrees of freedom

For a one-sample $t$ procedure, $\mathrm{df}=n-1$ . In a paired test, $n$ is the number of differences, which equals the number of complete pairs.

$10$ students measured $2\text{ times}$ supply $10$ differences, so $\mathrm{df}=9$ .

04 · Match the method to the data

Choose the test and parameter

First decide whether you are studying a quantitative measurement or a categorical outcome. Then decide whether the observations form $1$ sample, paired measurements or independent groups.

Visual guide 1: identify the structure before choosing a method

1

sample $1$ quantitative mean

Compare a population mean with a benchmark. Unknown $\sigma$ → one-sample $t$ -test.

Matched pairs $1$ difference per pair

Subtract within each pair, then test the population mean of that difference list.

Another structureReconsider the method

Independent groups or a proportion need a different procedure.

$2$ measurements do not automatically mean $2$ independent samples. The relationship between observations determines whether a paired analysis is appropriate.

Define a parameter in a full sentence: “ $\mu$ is the mean fill volume, in mL, of all bottles in this lot.” Include the mean, response measurement and target population.

For pairs: “ $\mu_d$ is the mean before-minus-after completion-time difference, in minutes, for students in this academy performing these comparable tasks.” Include the order of subtraction.

Keep methods distinct: A one-sample $t$ -test answers a mean question when $\sigma$ is unknown. A proportion test answers a question about the fraction with a categorical outcome. A two-sample mean test handles $2$ independent groups; it is developed later in Unit 4.

Large $n$ does not require switching an unknown- $\sigma$ mean test to $z$ . A $t$ procedure still uses $s$ to estimate the unknown population $\operatorname{SD}$ . If $\sigma$ were genuinely known, a different setup would be relevant; this topic focuses on unknown $\sigma$ .

05 · Translate the question

Write the hypotheses

State the null as $H_0:\mu=\mu_0$ . Translate the research question into one of the listed alternatives:

The question determines the alternative, before examining the sample result.
Question wording	Alternative	Direction
Lower, less, below, under	$H_a:\mu\lt\mu_0$	One-sided, lower direction
Higher, more, above, exceeds	$H_a:\mu\gt\mu_0$	One-sided, upper direction
Different, changed, not equal	$H_a:\mu\ne\mu_0$	Two-sided, either direction

Visual guide 2:

3

questions,

1

benchmark

      EXCELLENT  
   Based on 51 reviews 
    
 
 
 
   
 
 
       Posted on Google  
   
   0000003998 : Abdad Alam Shamim Alam 
 
 
 
 Trustindex verifies that the original source of the review is Google. This is the bestest institution I have ever come to and I love it very much.
  
 
      Posted on Google  
   
   Aliki S 
 
 
 
 Trustindex verifies that the original source of the review is Google. 
  
 
      Posted on Google  
   
   Ali 
 
 
 
 Trustindex verifies that the original source of the review is Google. Sly is him
  
 
      Posted on Google  
   
   Jitendra Kumar Kumawat 
 
 
 
 Trustindex verifies that the original source of the review is Google. 
  
 
      Posted on Google  
   
   Raahil Hasan 
 
 
 
 Trustindex verifies that the original source of the review is Google. 
  
 
      Posted on Google  
   
   Tina Mudarres 
 
 
 
 Trustindex verifies that the original source of the review is Google. The classes are amazing and my child learnt so much
  
 
      Posted on Google  
   
   alisha gadoya 
 
 
 
 Trustindex verifies that the original source of the review is Google. I have experienced a lot of good things, it has taught me so many things and from B/Cs I have gone to an A! This is wonderful and Mavish’s class is awesome.
  
 
      Posted on Google  
   
   smasher 123 
 
 
 
 Trustindex verifies that the original source of the review is Google. It’s sensational my kid went. First he only would get C now it’s all A’s really good and recommend
  
 
      Posted on Google  
   
   Mosa Al- Samaraie 
 
 
 
 Trustindex verifies that the original source of the review is Google. AMAZING PRICES GREAT TEACHERS STRAIGHT FORWARD LEARNING STEADY PACE IN TUTORING
  
 
      Posted on Google  
   
   Cael Dagnelie 
 
 
 
 Trustindex verifies that the original source of the review is Google. I had a great experience learning math with Ms. Mavish. She explains complex topics in a very clear and simple way, which made it easier for me to understand and enjoy the subject. Her patience and dedication really stood out, and she always made sure that everyone in the class was keeping up. I especially appreciated how approachable she was — I never felt afraid to ask questions, and she was always willing to help. Thanks to her teaching, my confidence in math has grown a lot. I’m really thankful for the effort she puts into every lesson!

NUM8ERS is one of finest tutoring institutes in UAE, Located in Al Barsha 1, Dubai. Close to DUBAI AMERICAN ACADEMY (DAA) & AMERICAN SCHOOL OF DUBAI (ASD).

Setting Up a Test for a Population Mean or Population Mean Difference

By the end of this lesson, you should be able to:

The concept in 60 seconds

A bottle-filling investigation