REPLY to WK 4-1 JuRi (100 words and 1 reference)
Type I error or a ‘false alarm’, occurs if a researcher rejects a null hypothesis that is true (Witte, 2017). An example of a type 1 is an innocent person being sentenced to prison in a court room. On the other hand, a type 2 error or ‘a miss’, occurs if a researcher fails to reject a null hypothesis that is actually false (Witte, 2017). An example of a type 2 error is the court room saying that the defender did not commit the crime, when they really did. Reducing the the significance level will help reduce the probability of a type 1 error (Witte, 2017). Increasing the sample size or choosing an alternative value from the parameter that is further from the null value will help reduce type 2 errors (Witte, 2017). An example of a testing both having a type 1 and 2 errors is if someone
REPLY to WK 4-1 JoRo (100 words and 1 reference)
Type I error occurs when a statistical test or hypothesis incorrectly rejects the null hypothesis when it is actually true (Witte & Witte, 2017). It is the incorrect rejection of a true hypothesis; often referred to as a “false positive”. An example would be a drug company trial testing the effectiveness of a new drug. The null hypothesis states that the drug is ineffective while the alternative hypothesis suggests that the drug is effective. If the company concludes that the drug is effective based on the trial results, but in reality, it is not, then it would be a Type I error. Type II error is when a statistical test or hypothesis fails to reject the null hypothesis when it is actually false (Witte & Witte, 2017). Continuing example: assume the same drug company concludes that the new drug is ineffective and fails to bring it to market. In reality the drug happens to be effective. This would mean the company has committed a Type II error by failing to reject the null hypothesis when it should have. In order to reduce both types of errors researchers may increase sample size—a larger sample size can improve the power of a statistical test, thereby reducing the risk of both Type I and Type II errors or—consider larger effect sizes for detection which would result in a decrease in Type II errors. Ensuring that the sample size is adequate to detect meaningful differences is essential.
REPLY to WK 4-2 ReCa
A significance test is used to test a hypothesis. Testing a hypothesis to find the signifycance level can determine if the null hypothesis should be rejected or if the hypothesis has a high probability of being correct (APA Dictionary, 2023). There are 3 significance testing levels, .01, .05 and .10. The lower the significance level, the less likely you are to reject the hypothesis. In my opinion when news articles fail to report the significance level of the study they do not intend on fully educating their audience. They are giving information that their audience can understand and will not question whether the information shared is accurate or not. If they shared the significance level they would have to explain in laments terms what it means and how the test was conducted including the hypothesis and the reason for rejection or not. The general public more than often does not understand the levels of significance and may interpret it wrong if not explained correctly. If this is a peer reviewed source in a medical journal etc., the explanation of the significance level would be expected as the readers understand it and would not be as accepting of it missing this information. If it’s in a general news article for the general public the news can get away with bare minimum information because people would tend to take it at face value.
