Explore BrainMass
Share

# Hypothesis Testing

Hypothesis testing is an integral topic in statistics and is conducted to test a population parameter or assertion for one or multiple samples. These hypotheses are generally based off of a theory or argument and tested using data collected from the samples.

The first step in hypothesis testing involves setting up the null and alternative hypotheses. The null hypothesis is written first and just as the name implies, this hypothesis claims that there is no difference or relationship between the observations being tested. Thus, the null hypothesis when written out numerically, always involves an equals (=) symbol.

The alternative hypothesis is written next and is the exact opposite of the null statement. Depending on the claim being tested, the alternative hypothesis can either propose a relationship in which one observation is greater than the other, less than the other, simply not equal to the other or a combination of those possibilities.

The alternative hypothesis is important for establishing whether the hypothesis test is characteristic of a left-tailed test, right-tailed test or 2-tailed test. The diagram below displays the probability density function for each type of test.

Figure 1.  In this case, the claim is based from mean values. The null hypothesis is denoted by μ and the alternative hypothesis is represented by μo. The left hand corner diagram represents the left-tailed test, the figure in the middle is representative of the two-tailed test and the distribution curve on the right shows the claim for a right-tailed test. Please view the references for the image credit.

Once the hypotheses are constructed, the next step involves setting the criterion for testing, which essentially means the significance level. Generally the significance level is set at 5%, but it is not restricted to this value. The significance value is dependent on the situation being investigated. Note, that in a two-tailed test, this 5% significance level is shared by both tails, so each tail represents a significance level of 2.5%.

The final steps involve computing the test statistic and then making a decision on the claim. There are two decisions which can be made: the null hypothesis can either be rejected or a researcher can fail to reject it. This decision is based on the calculation of a p-value. If the p-value is less than or equal to 5%, the null hypothesis is rejected.

References:

Image Credit: Shafer, D.S. and Zhang, Z. (2014). Small Sample Tests for a Population Mean. Retrieved from: http://2012books.lardbucket.org/books/beginning-statistics/s12-04-small-sample-tests-for-a-popul.html

## Categories within Hypothesis Testing

### Frequentist Inference

Solutions: 1,757

In the study of statistics, frequentist inference represents one of the main schools of thought in discussing the frequency and/or proportion of data from which conclusions are drawn.

### Parametric Tests

Solutions: 1,084

Parametric tests are statistical tests based on a set of assumptions about the probability distribution of a data set and this allows for a researcher to make inferences on the parameters of interest.

### Nonparametric Tests

Solutions: 919

Contrary to parametric tests, nonparametric tests do not make assumptions about the parameters of the probability distribution and are considered to be distribution-free tests.

### Type I and Type II Errors

Solutions: 129

Type I and Type II errors are measures which are directly linked with hypothesis testing and result from a researcher assessing the claim being tested inaccurately.

### Confidence Intervals, Hypothesis Testing, Sample Size

Comparing Two Population Means In 2004, the Mars Candy Company stated on their Web site that their 1.69 oz. package of M&M's contained an average of at least 53 candies and that 14% of all M&Ms manufactured were yellow. Students in a Math 1040 statistics class at the Salt Lake Community College were asked to determine if the

### Hypothesis Test Questions

1 Errors & Significance A STAT 200 instructor wants to compare the final exam scores of students taught using two different curricula. She obtains a sample of 500 students. She randomly assigns 250 students to a traditional curriculum and 250 students to a new curriculum. It is hypothesized that the new curriculum will lead t

### STA108 Chapters 16,17: General Statistics

1. A level 0.95 confidence interval is a. any interval with margin of error ± 0.95. b. a range of values computed from sample data that will contain the true value of the parameter of interest 95% of the time. c. a range of values with margin of error ± 0.95, which is also correct 95% of the time. d. a range of values comp

### PPG & MPG: correlation coefficient, competing hypothesis, value

The following table lists the National Basketball Association's leading scorers, their average points per game (PPG), and their average minutes per game (MPG) for 2008; the data are also available on the text website, labeled Points. PPG MPG D. Wade 30.2 38.6 L. James 28.4 37.7 K. Bryant 26.8 36.1 D. Nowitzki 25.9 37.3

### Confidence Interval and Tests of Hypotheses

1. The CDC estimated that influenza vaccination reduced the risk of flu illness in 600 of 1000 randomly sampled individuals during seasons when most circulating flu viruses are well-matched to the flu vaccine. Construct the 97% confidence interval for the population proportion. State your conditions, interpret your interval, and

### Inference for Two Samples

1. Market researchers working for a large fitness company want to know if male and female patrons differ in terms of how many hours per week they spend at the gym. At their gym, customers have ID cards that they use to sign in and out of the gym, so the researchers are able to obtain data concerning how long each customer was i

### EPA and Statistics self-efficacy

1. The Environmental Protection Agency (EPA) is testing tap water in one community. If they find evidence that the average level of lead is greater than 15 parts per billion (ppb) they will order that the water be turned off and that residents be given free bottled water until the problem is fixed. A. State the null and altern

### Chi-Squared and Regression

A pharmaceutical company develops a new drug that helps obese patients who are not currently diabetic avoid contraction of diabetes. The conduct conducts a trial splitting a sample of 135 patients into two groups, a drug group and a placebo group. They then monitored the patients for 5 years and noted how many patients in each g

### Z-Tests and T-Tests

The percent of hospitals in Region 3 of the northwestern United States that reached maximum occupancy at least once in January 2015 was 82%. The regional administration wants to conduct a study to determine if the rate has changed. Before conducting the study, she assumes that the average rate is still 82%. She calls a sample of

### Parameter and the population of interest for the hypothesis test

Eleven percent of the products produced by an industrial process over the past several months fail to conform to specifications. The company modifies the process to reduce the rate of non-conformities. In a trial run, the modified process produces 16 nonconforming items out of a total of 300 produced. Do these results demonstrat

### Counting, distribution, paired sample and bi-variate data

If you toss a die twice, what is the sample space if you want to restrict the results to only the ones where adding the results of the two throws together equals seven? For example, if you get 1 on the first toss and 6 on the second toss these two tosses meet the criteria. Which statistics distributions are based on the norma

### Correlation Descriptive Analysis

I have started building the body of the paper . See attached question and draft paper . I will like you to recommend - The descriptive analysis to run - The Statistics tests to run - the type of data i should provide from my statistics to run the tests - once i provide the data then we can move to analysis - See also

### Perform a hypothesis testing for one mean using t distribution

Conduct a one-tailed hypothesis test given the information below. A manufacturer wants to increase the shelf life of a line of cake mixes. Past records indicate that the average shelf life of the mix is 116 days. After a revised mix has been developed, a sample of nine boxes of cake mix had a mean of 117.333 and a standard de

### hypothesis testing for two means using two paired samples t test

A company is researching the effectiveness of a new website design to decrease the time to access a website. Five website users were randomly selected and their times (in seconds) to access the website with the old and new designs were recorded. The results follow: User Old Website Design New Website Design A 30 15 B 45 20

### Conduct a two-tailed hypothesis test given the information below.

ABC producing precision ball bearings. It is important that the diameters be as close as possible to an industry standard. The output from each process is sampled and the average error from the industry standard is measured in millimeters. The results are presented next. Process A Process B Sample Mean

### Hypothesis testing using either t or normal distribution

1. A sample of 49 observations is selected from a normal population. The sample mean is 21, and the population standard deviation is 4. Conduct the following test of hypothesis using the 0.05 significance level. H0 : μ ≤ 20 H1 : μ > 20 a. Is this a one- or two-tailed test? "One-tailed"-the alternate hypo

### Percentiles and Hypothesis Testing with Z-Tests

HOMEWORK 6 Percentiles and Hypothesis Testing with Z-Tests When submitting this file, be sure the filename includes your full name, course and section. Example: HW6_JohnDoe_354B01 Be sure you have reviewed this module/week's lesson and presentations along with the practice data analysis before proceeding to the homework exer

### Chi-Square Tests and Linear Regression

5. At a school pep rally, a group of sophomore students organized a free raffle for prizes. They claimed that they put the names of all the students in the school in the basket and that they randomly drew 36 names out of this basket. Of the prize winners, 6 were freshmen, 14 were sophomores, 9 were juniors, and 7 were seniors. T

### Stock Analyst Statistics

A stock analyst wants to determine whether there is a difference in the mean rate of return for three types of stock: utility, retail, and banking stocks. The following output is obtained: Picture a. http://ezto.mheducation.com/13252703214728584664.tp4?REQUEST=SHOWmedia&media=lind16e.png Using the .05 level of significan

### BMI, Caffeine Consumption, Crossover

Week 4 Problems—30 points 10 points per problem. • The mean body mass index (BMI) for boys age 12 is 23.6. An investigator wants to test if the BMI is higher in 12-year-old boys living in New York City. How many boys are needed to ensure that a two-sided test of hypothesis has 80% power to detect a difference in BMI

### Excel Cake Mix

Copy the Assignment Data (i.e., cake mix data) to Sheet 1, cells A1:P32, of a new Excel workbook. Select the most appropriate hypothesis test described in Sections 4.3 and 4.5 to evaluate the following null hypothesis using this worksheet: There is no difference in mean grams of sugar between name brands and generic brands o

### Hypothesis Testing using Alpha

NFL2000 Using these data, perform an appropriate test of hypothesis, at the alpha = .05 level of significance, to determine whether compensation for players in the NFL is dependent on the team for which they play. What is your conclusion? NFL salaries and bonuses 2000 season

### Single-sample t-test

HOMEWORK 8 Single-Sample T-Test When submitting this file, be sure the filename includes your full name, course and section. Example: HW8_JohnDoe_354B01 ________________________________________ Part I: Concepts Questions 1-3 ________________________________________ These questions are based on the Nolan and Heinzen re

### Biostat Case Study

There are seven questions at the end of this article. This assignment is worth 80 points. Place answers in this table: 1. 2. 3. 4. 5. 6. 7. BIOSTAT Case Study: Tests of Association for Categorical Data LEARNING OBJECTIVES At the completion of this Case Study, participants should be able to:  Compa

### Conducting a one-way ANOVA

A medical researcher wants to determine whether there is a difference in the mean length of time it takes three types of pain relievers to provide relief from headache pain. Several headache sufferers are randomly selected and given one of three medications. Each headache sufferer records the time (in minutes) it takes the medic

### Statistical Analysis Techniques: ANOVA and Post-Hoc Tests

A market researcher is interested in knowing the type of training that works best for DVD users. Thirty consumers are randomly selected from a population of known DVD owners (i.e., users). Ten users are trained by giving them the DVD user's manual and allowing them to read it. Another ten users are trained from a 30 minute DVD u

### Problem on Confidence Interval of Proportion

A study was carried out to understand the amount of time put in by the students in three management institutes. One hundred students from different institutions were surveyed. Information regarding the hours put in by the students per week is summarized in the following contingency table: Institution <= 40 hours/week between

### Construting a 95% confidence interval for the population mean

Question 1: Clothing for runners. Your company sells exercise clothing and equipment on the Internet. To design the clothing, you collect data on the physical characteristics of your different types of customers. Here are the weights for a sample of 24 male runners. Assume that that these runners can be viewed as a random sample

### Design of Experiments: Variables, Blocking, Factors, and Effects

Explain the difference between multiple independent variables and multiple levels of independent variables. Which is better? What is blocking and how does it reduce "noise"? What is a disadvantage of blocking? What is a factor? How can the use of factors benefit a design? Explain main effects and interaction effects. How does a

### Assumptions for a Statistical Test

Why does it matter whether the assumptions required for a statistical test are met?