The p value of 0.05 is the most often used across all disciplines. The p-value of 0.05 is most often used to compute 95% confidence level, which is the interval estimate of the poulation paramter based on 0.05 level of significance . Other frequently used o-values include 0.01 and 0.10.

You may find the following useful:

Article Statistical tests, P values, confidence intervals, and power...

Level of significance gives the probability of alpha error, i.e. the probability of rejecting the null hypothesis when it is true.

Confidence level, for example 95%, associated with a range of values implies that 95% of time , measured value of the random variable will lie in that range.

My Dear Muhammad I. hope you is good

approprite level of sig.is 0.05 in general and C.I is 95% ;but in the last years all we depended on software and we can see sig. level as output of computer this called p-value and reject Ho if p-value

Most of us would have used these terms and values in our statistical analysis and estimation. Both terms come from the Neyman-Pearson theory of hypothesis testing and both are related. They sound similar and thus are also confusing when used in practice. While the purpose of these two are invariably the same, there is a minor and important difference between these two terms conceptually, which makes them to inevitably devote an article to them.

Thanks to all respondents, please correct me, if

Level of significance (alpha) is 5%,

Confidence level (1-alpha) is 95%,

Can we use these two terms together? Or both used interchangeably? Please explain in simple words that when do we prefer using alpha than 1-alpha, vice versa.

Interesting. It is 95 level of confidence, .05 level of signifance.

In my own opinion, I think we can't say 95 level of significace and .05 level of confidence. It will mean diffirently.

There is difference between alpha and 1-alpha ; because alpha is certeria of signficant or probability of reject Ho when Ho is true (type one error)but 1-alpha is percentage that mean of puplation in this interval

and we seclected alpha 0.05. Or sometimes 0.01 and now by software we get p-value directly

Thank you so much Fadhil_Abidi

Thank you; I hope all success for all your works

Best Regards

The significance level indicates how much type I error you are willing to tolerate in your conclusions. Because you reject the null hypothesis when the p-value is low enough, a pre-defined significance level will place an upper limit to the probability of observing an observed sample statistic, given the fact that the null hypothesis holds true.

The confidence level is the complement of the significance level, and represents the probability that a given interval built around the point estimate (sample statistic) contains the unknown population parameter of interest.

following

Confidence interval is nothing but a range of values within which the unknown population parameter lies. It expresses certainty .95% Confidence interval means the certainty is 95%.

Meanwhile, the significance level is different. 5% ( or 0.05 ) significance level implies that there is 5% chance of rejecting the null hypothesis when it is true.Its value is generally less than 5%.

5% significance level corresponds to ( 100-5 =95 ) % confidence interval.

Probabilistic approach is very much an applied concept in Engineering. There is nothing like 100% so we use confidence intervals. For example, we randomly choose various specimens ( samples ) out of total populations and test them. If 95% ( or confidence interval ) of the samples meet the criteria, we accept the null hypothesis that the entire population complies to the specified criteria. This directly means that 5% error ( or significance level ) is tolerated.

Thank you Nirmala S.V.S.G and Kundan Raj Parajulee.

Above you are getting the text-book definitions. Yet you should be aware that the actual choice of alpha (i.e. the level of significance, also called the p-level) is arbitrary. Humans tend to be blind to hidden biases in their experiments. Most of these biases deflate estimated alpha values & increase the likelihood of false positives. This is why many experiments are "not reproducible". In the hard sciences, like physics, this is a recognized fact, & they compensate by using the 5-sigma rule rather than the 2-sigma rule (i.e. the 5% rule). It is my experience that the "softer the science" the more generous the alpha level chosen (mostly the 5% or 2-sigma level). I personally recommend using a minimum of the 3-sigma rule (the 1% level). This will help compensate for hidden biases.

@Jack Lothian thank you so much sir, we actually need some explanation in the discussion rather sharing links or definitions from books.

we use "level of significance alpha" in test hypothesis we selecte 0.05 but in medical work we select 0.01 ,Now in software we see p-value output and compare p-value with alpha (reject Ho if p-value

@Fadhil Abdulabbas Abidi, thank you very much sir.