click below
click below
Normal Size Small Size show me how
Ch10_Hyp_Testing
Hypothesis Tests Regarding a Parameter
Question | Answer |
---|---|
What is a “hypothesis”? | A “hypothesis” is a statement regarding a characteristic of one or more populations. |
In statistics, what do we mean by “hypothesis testing”? | “Hypothesis testing” is a procedure, based on sample evidence and probability used to test statements regarding a characteristic of one or more populations. |
How is the “null hypothesis” used in hypothesis testing? | The “null hypothesis”, denoted H0, is a statement to be tested regarding a population parameter (statement of no change, no effect or no difference and is assumed true until evidence indicates otherwise. |
How is the “alternative hypothesis” used in hypothesis testing? | The “alternative hypothesis”, denoted H1, is a statement that we are trying to find evidence to support regarding the value of a population parameter. |
Suppose the hypothesis test regarding a population parameter is set up as follows: H0: parameter = some value H1: parameter ≠ some value What type of hypothesis test is this: left-tailed, right-tailed or two-tailed? | This is a Two-tailed hypothesis test. |
Suppose the hypothesis test regarding a population parameter is set up as follows: H0: parameter = some value H1: parameter < some value What type of hypothesis test is this: left-tailed, right-tailed or two-tailed? | This is a Left-tailed hypothesis test. |
Suppose the hypothesis test regarding a population parameter is set up as follows: H0: parameter = some value H1: parameter > some value What type of hypothesis test is this: left-tailed, right-tailed or two-tailed? | This is a Right-tailed hypothesis test. |
According to one public official, the average length of a 911-call in 2008 was 3.50 minutes. A researcher believes that the average length of a 911-call has increased since then. State the Null and Alternative hypothesis. | H0: µ = 3.5 H1: µ > 3.5 Right-tailed test |
A part produced using an old machine with a standard deviation of the diameter of 0.052 inches. The engineering department believes that the new machine can reduce the standard deviation of the part. State the Null and Alternative hypothesis. | H0: σ = 0.052 H1: σ < 0.052 Left-tailed test |
John makes the statement that 38% of American adults exercise on a regular basis. A researcher believes that the actual percentage is different. State the Null and Alternative hypothesis. | H0: p = 0.38 H1: p ≠ 0.38 Two-tailed test |
In statistics, what do we mean by a “Type I error”? | A Type I error is made when we reject the null hypothesis (H0) when the null hypothesis (H0) is true. |
What symbol is used for the probability of making a Type I error? | The symbol used for the probability of making a Type I error is α (“alpha”). |
In statistics, what do we mean by a “Type II error”? | A Type II error is made when we do not reject the null hypothesis (H0) when the alternative hypothesis (H1) is true. |
What symbol is used for the probability of making a Type II error? | The symbol used for the probability of making a Type II error is β (“beta”). |
In statistics, how is the term “level of significance used? | The “level of significance”, a, is the probability of making a Type I error. |
What is the relationship between the probability of a Type I error and the probability of a Type II error? | As the probability of a Type I error increases, the probability of a Type II error decreases, and vice-versa. |
Is it proper to say that we “accept” the Null Hypothesis? | No. We never “accept” the null hypothesis. Rather, we say that we “do not reject” the null hypothesis. This is just like the court system. We never declare a defendant “innocent”, but rather say the defendant is “not guilty”. |
Suppose a researcher wants to test the claim that the average amount of the over-limit fee on credit cards in the U.S. is $39 and sets up the following hypotheses: H0: µ = 39 H1: µ > 39. Explain what it would mean to make a Type I error. | A “Type I” error is made if the researcher concludes that the average amount of the over-limit fee on credit cards in the U.S. greater than $39 when the true average amount of the over-limit fee on credit cards in the U.S. is $39. |
Suppose a researcher wants to test the claim that the average amount of the over-limit fee on credit cards in the U.S. is $39 and sets up the following hypotheses: H0: µ = 39 H1: µ > 39. Explain what it would mean to make a Type II error. | A “Type II” error is made if the researcher concludes that the average amount of the over-limit fee on credit cards in the U.S. is $39 when the true average amount of the over-limit fee on credit cards in the U.S. is greater than $39. |
Before testing hypotheses regarding a population mean assuming the population standard deviation is known, what requirements must be met? | The following requirements must first be satisfied: 1. A simple random sample is obtained. 2. The population from which the sample is drawn is normally distributed, OR 3. The sample size is large (n≥30). |
In hypothesis testing regarding a population mean assuming the population standard deviation is known, what is the formula for calculating the test statistic? | Because the population standard deviation is “known”, we would use a “Z-Interval”. The formula for the test statistic would be: z0 = (x-bar – µ0)/(σ/sqrt(n)) |
Using the classical method of hypothesis testing regarding a population mean, assuming the population standard deviation is known, for a “two-tailed test”, how are critical values calculated? | Since σ is known we are conducting a “Z-Test” with two-tails; so there will be two critical values: Using the TI-83/84 we have: z(α/2)= invNorm(1 – α/2, 0, 1) – z(α/2) = invNorm(α/2, 0, 1). |
How are the critical z-values, z(α/2) and – z(α/2), used to test the hypotheses for a two-tailed test? | For a two-tailed test, the Null hypothesis is REJECTED if the test statistic (z0) is in the either tail (the CRITICAL REGIONS). That is, if z0 < – z(α/2) OR z0 > z(α/2). Otherwise, we DO NOT REJECT the Null hypothesis. |
Using the classical method of hypothesis testing regarding a population mean, assuming the population standard deviation is known, for a “left-tailed test”, how is the critical value calculated? | Assuming the population standard deviation is known, we are conducting a “Z-Test”. Since this is a left-tailed test, there will be one critical value: – z(α). For the TI-83/84, use: invNorm(α, 0, 1) to find the critical Z value, –z(α). |
How is the critical z-value,– z(α), used to test the hypotheses for a left -tailed test? | For a left-tailed test, the Null hypothesis is REJECTED if the test statistic (z0) is in the left tail (the CRITICAL REGION). That is, if z0 < – z(α). Otherwise, we DO NOT REJECT the Null hypothesis. |
Using the classical method of hypothesis testing regarding a population mean, assuming the population standard deviation is known, for a “Right-tailed test”, how is the critical value calculated? | Assuming the population standard deviation is known, we are conducting a “Z-Test”. Since this is a right-tailed test, there will be one critical value: z(α). For the TI-83/84, use: invNorm(1 – α, 0, 1) to find the critical Z value, z(α). |
How is the critical z-value, z(α), used to test the hypotheses for a right -tailed test? | For a right-tailed test, the Null hypothesis is REJECTED if the test statistic (z0) is in the right tail (the CRITICAL REGION). That is, if z0 > z(α). Otherwise, we DO NOT REJECT the Null hypothesis. |
What does the “p-value” represent when you are testing a hypothesis concerning a population mean, µ? | When a hypothesis concerning a population mean, µ, is tested, the “p-value” is the probability of obtaining a sample mean at least as extreme as the value of “x-bar” observed if the value assumed by the null hypothesis were true. |
Under the p-value approach, how is the p-value used to test the hypotheses? | Once the p-value is calculated, the decision is straightforward: 1) If the p-value < α, REJECT the Null hypothesis; 2) If the p-value ≥ α, DO NOT REJECT the Null Hypotheses. |
Suppose you are performing a left-tailed test, where the population standard deviation is known, and the test statistic is -1.25. What is the p-value? | We are using a Z-test because the population standard deviation is known. Since this is a left-tailed test, we are looking for the p-value = P(Z < -1.25). Using the TI-83/84, we have: normalcdf(-1E99, -1.25, 0, 1) = 0.1056 |
In hypothesis testing regarding a population mean assuming the population standard deviation is “unknown”, what is the formula for calculating the test statistic? | Because the population standard deviation is “unknown”, we would use a “T-Interval”. The formula for the test statistic would be: t0 = (x-bar – µ0)/(s/sqrt(n)) |
To test hypotheses regarding the population mean with s “unknown”, what requirements must be checked before performing the test? | The following must be checked first: 1. simple random sampling obtained. 2. The sample has no outliers, and the population is normally distributed, OR 3. The sample size is large (n≥30). Same as for a Z-test for a population mean. |
Using the classical method of hypothesis testing regarding a population mean, assuming the population standard deviation is “unknown”, for a two-tailed test, how are critical values calculated? | Since σ is “unknown”, we are conducting a “T-Test”, two-tailed test so there are two critical values with n – 1 degrees of freedom. For the TI-83/84, use: t(α/2) = invT(1 – α/2, n – 1. and – t(α/2) = invT(α/2, n – 1) |
How are the critical t-values, t(α/2) and – t(α/2), used to test the hypotheses for a two-tailed test? | For a two-tailed test, the Null hypothesis is REJECTED if the test statistic (t0) is in the either tail (the CRITICAL REGIONS). That is, if t0 < – t(α/2) OR t0 > t(α/2), for n – 1 degrees of freedom. Otherwise, we DO NOT REJECT the Null hypothesis. |
Using the classical method of hypothesis testing regarding a population mean, assuming the population standard deviation is “unknown”, for a left-tailed test, how is the critical value calculated? | Since σ is “unknown”, we are conducting a “T-Test”. Since this is a left-tailed test, there will be one critical value with n – 1 degrees of freedom: For the TI-83/84, use: – t(α) = invT(α, n – 1) to find the critical T value. |
How is the critical t-value,– t(α), used to test the hypotheses for a left -tailed test? | For a left-tailed test, the Null hypothesis is REJECTED if the test statistic (t0) is in the left tail (the CRITICAL REGION). That is, if t0 < – t(α) for n – 1 degrees of freedom. Otherwise, we DO NOT REJECT the Null hypothesis. |
Using the classical method of hypothesis testing regarding a population mean, assuming the population standard deviation is “unknown”, for a Right-tailed test, how is the critical value calculated? | Since σ is “unknown”, we are conducting a “T-Test”. Since this is a “right-tailed test”, there will be one critical value with n – 1 degrees of freedom. For the TI-83/84, use: t(α) = invT(1 – α, n – 1). |
How is the critical t-value, t(α), used to test the hypotheses for a right tailed test? | For a right-tailed test, the Null hypothesis is REJECTED if the test statistic (t0) is in the right tail (the CRITICAL REGION). That is, if t0 > t(α) for n – 1 degrees of freedom. Otherwise, we DO NOT REJECT the Null hypothesis. |
For the T-test, under the p-value approach, how is the p-value used to test the hypotheses? | Once the p-value is calculated, the decision is straightforward: 1) If the p-value < α, REJECT the Null hypothesis; 2) If the p-value ≥ α, DO NOT REJECT the Null Hypotheses. |
Suppose you are performing a left-tailed test, where the population standard deviation is “unknown”, and the test statistic is -1.25 and the sample size is n = 20. What is the p-value? | We are using a T-test because the population standard deviation is “unknown”. Since this is a left-tailed test, we are looking for the p-value = P(T < -1.25). Using the TI-83/84, we have: tcdf(-1E99, -1.25, 19) = 0.1132 |
Before performing a hypothesis test for a “population proportion”, p, what requirements must be met? | We must first check the following: 1. The sample is obtained by simple random sampling. 2. The sampled values are independent of each other, i.e., n ≤ 0.05N. 3. np0(1-p0) ≥ 10. |
How would the hypothesis test for a claimed population proportion, p, be performed? | The steps would be the same as those used for testing a hypothesis for a population mean, using the Z-interval. Either the “classical approach” or “p-value approach” may be used. However, the formula for the calculating the test-statistic is different. |
What formula is used for calculating the test-statistic when we are testing a hypothesis about population proportion? | The formula used for calculating the test-statistic when we are testing a hypothesis about population proportion is: Z0 = (p-hat – p0)/sqrt[p0(1 – p0)/n] |
A researcher claims that 48% of Americans are unhappy with politicians in Washington; in a poll 535 out of 1010 adults stated they are not happy with politicians in Washington. Calculate the value of the test statistic to be used in the hypothesis test. | First find P-hat = 535/1010 = 0.5297. Therefore, the test statistic is: Z0 = (0.5297 – 0.48)/sqrt[0.48(1 – 048)/1010] = 3.162 |
Based on the previous problem, if you want to test whether the percentage of Americans unhappy with politicians in Washington is more than the researcher claimed, using the p-value method, what conclusion would you reach? Use α= 0.05. | First, find p-value = P(Z > 3.162) = 0.000784; p-value (0.000784) < α = 0.05; REJECT H0. There is sufficient evidence to conclude the percentage of Americans unhappy with politicians in Washington is more than the 48% claimed by the researcher. |