Statistical Test of Hypotheses

Professor M. Kabir Department of Statistics, Jahangirnagar University

Hypothesis
A hypothesis may be defined is simply as a statement about one or more populations. The hypothesis is frequently concerned with the parameters of the populations about which the statement is made.

Hypothesis
A hospital administrator may hypothesize that an average length of stay of patient admitted to the hospital in five days; A public health nurse may hypothesize that a particular educational program will result in improved communication between nurse and the patient A physician may hypothesize that a certain drug will be effective in 90% of the cases for which it is used.

Hypothesis
By means of hypothesis testing one determines whether or not such statements are compatible with available data. Types of Hypotheses There are two types of hypotheses - Research hypotheses - Statistical hypotheses

Hypothesis
The research hypothesis is the conjecture or supposition that motivates the research. Research hypotheses lead directly to statistical hypotheses. Statistical hypotheses are hypotheses are stated in such a way that they may me evaluated by appropriate statistical techniques.

Hypothesis
Steps in Hypothesis Testing Data Assumptions Hypotheses Test Statistic Distribution of Test Statistic Decision Rule

Hypothesis
There are two statistical hypotheses involved in hypotheses testing. These are null hypotheses and alternative hypotheses. A null hypothesis specifies a hypothesized real value, or values for a parameter. It is denoted by the symbol Ho. The null hypothesis is sometimes referred to as a hypothesis of no difference, since it is a statement of agreement with conditions presumed to be true in the population of interest.

Hypothesis
An alternative hypothesis specifies a real value or range of values for a parameter that will be considered when the null hypothesis is rejected. The alternative hypothesis is a statement of what we will believe is true if our sample data cause us to reject the null hypothesis. Usually the alternative hypothesis and research hypothesis are the same, and in fact the two terms are used interchangeably. We shall designate alternative hypothesis by the symbol Ha.

Hypothesis
The test statistic is some statistic that may be computed from the data of the sample. The test statistic serves as a decision maker, since the decision to reject or not to reject the null hypothesis depends on the magnitude of the test statistic value What is rejection region? The rejection region consists of the set of values of a statistic for which the null hypothesis is rejected. The values of the boundaries of the region are called the critical values.

Hypothesis
What is type one error? A type I error occurs when the null hypothesis is rejected when in fact it is true. The significance level is the probability of a type one error when the null hypothesis is true. What is type II error? A type II error occurs when the null hypothesis is not rejected when it is false

Hypothesis
The power of a test is the probability of rejecting the null hypothesis when it is false. The probability of a type I error is denoted by , and the probability of a type II error is by . The power is defined as Power= 1 probability of type II error Power = 1- .

Hypothesis
Normal Test when population mean and variance is known

x x

/ n

Hypothesis
General Formula for Test Statistic The following is a general formula for a test statistic that will be applicable in many of the hypothesis tests discussed Test statistic = relevant statistichypothesized parameter/ standard error of the relevant statistic

Hypothesis
Distribution of Test Statistic
It has been pointed out that the key to statistical inference is the sampling distribution. x The distribution of test statistic z / n for example follows the standard normal distribution if the null hypothesis is true and the assumptions are met.

Hypothesis
Decision rule: The decision rule tells us to reject the null hypothesis if the value of the test statistic that we compute from our sample is one of the values in the rejection and to reject the null hypothesis if the computed value of the test statistic is one of the values in the nonrejection region.

Hypothesis
Significance level: The decision as to which values go into the rejection region and which ones go into the non rejection region is made on the basis of the desired level of significance, designated by . The term level of significance reflects the fact that hypothesis tests are sometimes called significance tests, and computed value of the test statistic that falls in the rejection region is said to be significant.

Hypothesis
The level of significance, specifies the area under the curve of the distribution of the test statistic that is above the values on the horizontal axis constituting the rejection region. Types of errors The error committed when a true null hypothesis is rejected is called type I error . The type II error is the error committed when a false null hypothesis is not rejected. The probability of committing type II is designated by

Hypothesis
Whenever we reject a null hypothesis there is always the concomitant risk of committing a type I error, rejecting a true null hypothesis. Whenever we fail to reject a null hypothesis the risk of falling to reject a false null hypothesis is always present.

Hypothesis
Statistical Decision The statistical decision consists of rejecting or of not rejecting the null hypothesis . It is rejected if the computed value of the test statistic falls in the rejection region, and is not rejected if the computed value of the test statistic falls in the non-rejection region. Conclusion: If Ho is rejected we conclude that Ha is true. If Ho is not rejected we conclude that Ho may be true.

Hypothesis
The p value is the smallest value of for which the null hypothesis can be rejected. For Z= -2.12 the p value is 0.034. The p value for a hypothesis testing is the probability of obtaining when Ho is true, a value of the test statistic as extreme or more extreme than the one actually computed. If p value is less than or equal to , we reject the hypothesis. If p value is greater than , we do not reject the hypothesis. We accept the hypothesis

Hypothesis
Hypothesis Accept Ho Reject Ho

Accept Ho

Correct

Type II error

Reject Ho

Type I error

Correct

Steps in Hypothesis Testing

Evaluate data Review assumption State hypothesis Select rest statistics Determine distribution of test statistic State decision rule Calculate test statistic Make statistical decision Do not reject Ho Reject Ho Conclude Ho may be true Calculate Ha is true

Hypothesis Testing: A Single Population Mean

We consider the testing of a hypothesis about a population mean fewer than three different conditions When sampling is from a normally distributed population of values with known variance When sampling is from a normally distributed population of values with unknown variance

Hypothesis Testing: A Single Population Mean

When sampling is from a normally distributed population and the population variance is known, the test statistic for testing Ho:
z x

/ n

EX. If random sample of size 10 is drawn from a normal population with mean and variance are respectively 27 and 20 respectively. Can we conclude the mean age of this population is different from 30 years?

Hypothesis Testing: A Single Population Mean

EX. If random sample of size 10 is drawn from a normal population with mean and variance are respectively 27 and 20 respectively. Can we conclude the mean age of this population is different from 30 years? Calculation of test statistic We have z= (27-30)/ 1.4142 = - 2.12 We reject hypothesis. We conclude that population mean is different from 30 years.

Hypothesis Testing: A Single Population Mean

Testing Ho by means of a confidence interval The 95% confidence interval of population mean is given by 27 plus-minus 1.96 Square root of 20/10 27+ 2.7718, 27-2.7718 The age lies between 29.77 to 24.23 years Since the interval does not include 30, we say 30 is not a candidate for the mean we are estimating and there fore population mean is not equal =30 and Ho is rejected.

Hypothesis Testing: A Single Population Mean

In general, when testing null hypothesis by means of a two sided confidence interval, we reject Ho at the level of significance if the hypothesized parameter is not contained with the 100 ( 1-) percent confidence interval. If the hypothesized parameter is contained within the interval, Ho cannot be rejected at the level of significance

Hypothesis Testing: A Single Population Mean

Sampling from a normally distributed population: Population Variance is unknown The test statistic for testing Ho: Population mean= = o is Statistic is sample mean- pop population / s/square root of n Example: Will we be able to conclude that the mean BMI for the population is 35 .

Hypothesis Testing: A Single Population Mean

Can we reject the hypothesis that population mean is equal to 35. Body Mass Index ( BMI) measurements for 14 male subjects are given below.
Subject 1 2 3 4 5 6 7 8 9 10 11 12 13 14 BMI 23 25 21 37 39 21 23 24 32 57 23 26 32 45

Hypothesis Testing: A Single Population Mean

Data: The data consist of BMI measurements on 14 subjects as given above. Assumptions: The 14 subjects constitute a simple random sample from a population of similar subjects. We assume that BMI measurements in this population are approximately normally distributed. Hypotheses: Population mean 35 Population is not equal to 35 x t Test statistic is with d.f is n-1, s/ n t= (30.5-35)/2.8434

x s/ n

Hypothesis Testing: A Single Population Mean

The calculated value of t = -1.58 With 13 degree of the value of t is 2.16 Since computed value of t is less than the table value. We accept the hypothesis. Based on the data the mean population from which the sample drawn may be 35.
Hypothesis Testing: population standard deviation is not known

x s/ n

If the population standard deviation is not known , the usual practice is to use the sample standard deviation as an estimate. The test statistic for testing Ho= = o,

Hypothesis Testing: A Single Population Mean

then , which when Ho is true , is s/ n distributed approximately as the standard normal distribution if n is large.
t x

x s/ n

Hypothesis Testing: A Single Population Mean

Ex. A study was conducted to describe the menopausal status , menopausal symptoms, energy expenditure, and aerobic fitness of healthy midlife women and to determine the relationship among these factors. The mean score of maximum oxygen uptake for a sample 242 was 33.3 with a standard deviation of 12.14. The researcher wishes to know if, on the basis of these data, one may conclude that the mean score for a population of such women is greater than 30

x s/ n

Hypothesis Testing: A Single Population Mean

Data Maximum oxygen uptake for 242 women with mean = 33.3 and s = 12.14 Assumptions: The data constitute a simple random from a population of healthy midlife women similar to those in the sample. Hypotheses Ho: grater than equal to 30 Ha: is greater than 30

x s/ n

Hypothesis Testing: A Single Population Mean

Data Maximum oxygen uptake for 242 women with mean = 33.3 and s = 12.14 Assumptions: The data constitute a simple random from a population of healthy midlife women similar to those in the sample. Hypotheses Ho: grater than equal to 30 Ha: is greater than 30

x s/ n

Hypothesis Testing: A Single Population Mean

Given the above test statistic We have z= ( 33.3-30)/0.7804= 3.3/0.7804 = 4.23 We reject the hypothesis since computed value is greater than table value. We conclude that the mean score for the sampled population is greater than 30.

x s/ n