The major downside is 30/900, over 3%, might be biased if you retest them. It is highly unlikely you got the survey 100% correct the first time and had no revisions. To pilot a survey, one interviews participants and analyzes data. So, the test-retest effect is a real concern.

While it is generally recommended that pilot testing should be conducted with a separate group of respondents from those who will participate in the final survey, there can be exceptions to this rule. Here are some considerations:

Advantages of Using Separate Pilot Test Respondents:

However, in some situations, it may be acceptable to use the same population for pilot testing, especially if it is logistically challenging to find a separate group. Here are considerations for conducting pilot testing with the same population:

In summary, while it is advisable to use a separate group for pilot testing to minimize bias and obtain unbiased feedback, there are situations where conducting the pilot test within the same population can be acceptable, as long as participants are informed and precautions are taken to address potential biases and improve the survey instrument based on the pilot test results.

Pilot testing respondents should be different from the final survey respondents for the accuracy of results.

It is not wise to do a pilot in your final data collection. Remember the purpose of the pilot is to make your data collection instrument reliable and valid. What if you have to change the structure of a question or the options for a question, the data collected might not be reliable.

It's generally good practice in survey research to conduct a pilot test with a separate group of participants who won't be part of the final survey. Using a separate group for the pilot test allows you to identify and rectify any issues before deploying the survey to your target population.

I do not see why not to include the participants of the pilot test, as long as the instrument does not have contradictory answers and some items have to be eliminated, but if all the answers indicate that such contradictions do not exist, I think there is no problem with including those who take the pilot test.

The respondents in the pilot testing should not be included in the final testing

Better is to separate the results well

That would not be the best as it may affect the results of the final survey

I would like to hear from all those who believe that the results of the pilot test should not be included in the final results as a single sample. I repeat as I said in my previous answer, if there are no contradictions in the answers of any item as to consider that it can be eliminated, please explain to me statistically how it would affect the final results

Generally, a pilot involves changing the instrument before using the three study proper. Test-retest bias is a common problem. Also, often one conducts interviews after a pilot with the sample, which can introduce bias if retested.

What is recommended is 3 pilot samples from 3 different educational establishments with characteristics similar to your accessible population.

Each pilot with a maximum of 10 analysis units. The purpose of piloting is to give reliability and validity to the data collection instrument.

An instrument is built in its first version, with 10 items based on the specific objectives that you have formulated. Then, with the observations and suggestions of the first pilot, a second version of the instrument is developed; With its corrections and suggestions, a third version is prepared that must be applied to the third pilot and with its suggestions and corrections, a fourth version of the instrument must be prepared that must be validated for less variability.

The fourth version of the instrument is the one that should be applied to your work sample.

Therefore, none of the pilots should participate as a sample because it biases the results, since they will tend to have perfect scores on the rating scale.

I wouldn't recommend it. Not only will it most likely be biased, but also raises issues of validity and/or reliability.

Pilot study sample cannot be counted for final study.Pilot testing is done away from your study site so that it will not influence your data collection. Therefore by theory pilot tested sample cannot be from the sample you wanted to do the research.

Including the participants who participated in the research in the pilot study in the research again may have a negative effect on the validity and reliability processes. For the final study, it would be more appropriate to have participants who have not encountered the data collection tool before.

NO they should be excluded in your final survey. Use slovin"s formula in determining your sample size..

If you want unbiased and trustworthy results then don’t. I have seen some researchers that may say “Sure, you can let your pilot testers take the final survey”. Which does not make sense to me because you don’t have any guarantee that they will give an honest fresh answer since they've seen the questions before. I don’t think is okay even if you don't have enough people or if the survey didn't change much after the pilot. The best is always to get new folks to avoid any bias from the earlier run. :)

They are not supposed to be part of the final survey. The reason being that they have a knowledge of what the research is all about and may give response that may be bias.

You shouldn’t include the data from the pilot in the final analysis.

They can; but they shouldn't. They would skew the result.