  home

bytes

tutorials

applied statistics

hypothesis testing and p values

# Hypothesis Testing and p-Values

Module - 6 Statistical Inference
Hypothesis Testing and p-Values

Overview

The article discusses the concepts of hypothesis testing and p-values in statistics. It explains the process of hypothesis testing and how p-values help in decision-making. The article also explores one-tailed and two-tailed tests, confidence intervals, and the relationship between hypothesis testing and confidence intervals.

Introduction to Hypothesis Testing

Hypothesis testing is a statistical technique that helps us make decisions about a population based on a sample of data. It involves making an assumption, called a hypothesis, about the population parameter(s) of interest and then using data to test whether that assumption is supported or contradicted. The two main types of hypotheses are the null hypothesis and the alternative hypothesis.

Null and Alternative Hypotheses

The null hypothesis, denoted by H0, is the assumption that there is no significant difference between a parameter of interest and a given value. The alternative hypothesis, denoted by Ha, is the opposite of the null hypothesis and represents the possibility of a significant difference between the parameter of interest and the given value.

For example, suppose we are interested in testing whether the average height of a population is equal to 170 cm. The null hypothesis would be that the average height is equal to 170 cm (H0: μ = 170), and the alternative hypothesis would be that the average height is not equal to 170 cm (Ha: μ ≠ 170).

Significance Level and Critical Values

The significance level, denoted by α, is the probability of rejecting the null hypothesis when it is actually true. It is often set to 0.05 or 0.01, depending on the level of confidence desired. The critical value is the value that separates the rejection and non-rejection regions.

For example, if we set α = 0.05 and perform a two-tailed test, the critical values would be -1.96 and 1.96. If the calculated test statistic falls within the rejection region (i.e., outside of these critical values), we reject the null hypothesis. If it falls within the non-rejection region, we fail to reject the null hypothesis.

Calculating p-Values

The p-value is the likelihood of getting a test measurement as extraordinary as, or more extraordinary than, the watched test measurement, expecting the invalid theory is genuine. It could be a degree of the quality of prove against the invalid theory.

For case, in case we perform a two-tailed test and calculate a test measurement of -2.5, the p-value would be the zone beneath the tail of the standard normal conveyance past -2.5 and 2.5. This could be calculated utilizing measurable program or a table of the standard typical dispersion. A p-value less than or rise to to the importance level demonstrates that we dismiss the invalid theory.

Interpreting p-Values

The p-value provides a measure of the strength of evidence against the null hypothesis. A small p-value (less than the significance level) indicates that the data is unlikely to have occurred by chance alone, and we have evidence to reject the null hypothesis. A large p-value indicates that the data is consistent with the null hypothesis, and we fail to reject it. One-Tailed vs. Two-Tailed Test

In a one-tailed test, we are testing the speculation in one direction as it were, either more prominent than or less than the hypothesized esteem. In a two-tailed test, we are testing the theory in both headings, either more noteworthy than or less than, but not break even with to the hypothesized esteem. The choice of a one-tailed or two-tailed test depends on the inquire about address and the theory being tried.

Confidence Intervals and Hypothesis Testing

A confidence interval is an gauge of the run of values that the populace parameter is likely to drop inside, given the test information and a certain level of certainty. Confidence intervals are utilized in theory testing to assist decide whether a invalid speculation ought to be rejected or not. In case the certainty interim for a populace parameter does not incorporate the hypothesized esteem, at that point the invalid theory can be rejected at the given level of centrality.

For case, on the off chance that a 95% confidence interval for the population mean height is (168, 172) cm, and the invalid speculation is that the cruel stature is break even with to 170 cm, at that point ready to dismiss the invalid speculation since the hypothesized esteem isn't inside the certainty interim.

Another way to perform hypothesis testing is by utilizing p-values. A p-value is a likelihood of getting a test measurement as extraordinary or more extraordinary than the watched esteem, expecting that the invalid theory is genuine. The p-value is compared to the importance level (α) to decide whether the invalid theory ought to be rejected or not. In the event that the p-value is less than the importance level, at that point the invalid theory can be rejected.

For case, on the off chance that the p-value is 0.03 and the importance level is 0.05, at that point the invalid speculation can be rejected since the p-value is less than the importance level.

Conclusion

Hypothesis testing and p-values are vital devices in factual investigation for deciding the probability of a given speculation being genuine or wrong. One-tailed and two-tailed tests are utilized to test speculations completely different headings, whereas certainty interims give gauges of the likely run of values for a populace parameter. P-values are utilized to decide whether a invalid speculation ought to be rejected or not based on the watched information and a given importance level.

Key takeaways

1. Hypothesis testing may be a statistical strategy utilized to decide whether a speculation almost a populace parameter is backed by the test information.
2. The p-value is the likelihood of getting a test measurement as extraordinary as the one watched or more extraordinary, accepting the invalid speculation is genuine.
3. The noteworthiness level, alpha, is the likelihood of dismissing the invalid theory when it is really genuine, and it is ordinarily set at 0.05 or 0.01.
4. Confidence intervals are utilized to appraise the extend of values that the populace parameter is likely to drop in, and theory testing can be utilized to decide whether a given esteem is inside that extend or not.
5. It is important to carefully consider the null and alternative hypotheses, choose an appropriate test statistic and significance level, and interpret the results in the context of the problem at hand.

Quiz

1. What is the alternative hypothesis in a one-tailed test?

a) Ha: μ = 170

b) Ha: μ > 170 or Ha: μ < 170

c) Ha: μ ≠ 170

d) Ha: σ > 170 or Ha: σ < 170

Answer: b) Ha: μ > 170 or Ha: μ < 170

2. What is the critical value used in hypothesis testing?

a) The probability of making a Type II error

b) The point estimate of the population parameter

c) The value used to determine if the test statistic is significant

d) The confidence level of the test

Answer: c) The value used to determine if the test statistic is significant

3. What is the relationship between confidence intervals and hypothesis testing?

a) Confidence intervals and hypothesis testing use different methods to estimate population parameters

b) Confidence intervals and hypothesis testing use the same methods to estimate population parameters

c) Confidence intervals and hypothesis testing are not related

d) Confidence intervals and hypothesis testing are used for different types of data

Answer: b) Confidence intervals and hypothesis testing use the same methods to estimate population parameters

4. What is the p-value in hypothesis testing?

a) The probability of rejecting the null hypothesis when it is true

b) The probability of rejecting the null hypothesis when it is false

c) The probability of making a Type I error

d) The probability of making a Type II error

Answer: a) The probability of rejecting the null hypothesis when it is true

Related Tutorials Python Tutorial  2007 SQL Tutorial  737 Data Science Tutorial  1178 MLOps  552

Related Articles How does Zomato use Machine Learning?  8 mins  3581 Here Is How Ai Is Changing the World of Sports Forever!  11 mins  2378 How Machine Learning is Revolutionizing Customer Credit Risk Management  5 mins  3109 Implementation of Credit Risk Using ML  9 mins  2096 How Netflix Uses ML & AI For Better Recommendation for Users  9 mins  2792 Why do we always take p-value as 5%?  7 mins  4644

AlmaBetter’s curriculum is the best curriculum available online. AlmaBetter’s program is engaging, comprehensive, and student-centered. If you are honestly interested in Data Science, you cannot ask for a better platform than AlmaBetter. Kamya Malhotra
Statistical Analyst
Fast forward your career in tech with AlmaBetter

Vikash SrivastavaCo-founder & CPTO AlmaBetter Related Tutorials to watch  Made with  in Bengaluru, India