probability = the percentile you’re interested in converting. It turns out that a percentile of 0.85 corresponds to a z-score of roughly 1.036. In plain English, this means a data value located at the 85th percentile in a dataset has a z-score of 1.036. Z-scores can take on any value between negative infinity and infinity.
https://www.statology.org/convert-z-scores-percentiles-excel/
Using the data set below, here's an example of calculating the 60th percentile: Rank the values in the data set in order from smallest to largest, as shown below. Calculate the index. ... Round the index to the nearest whole number (5). To calculate percentile according to the 'greater than' method, count the values in your data set from smallest to largest until you reach the number ranked 5th, as determined in ... More items...
https://www.reference.com/world-view/calculate-z-score-percentile-1e205d9be0396641
Consult a z-score chart Z-score charts are available online and in any Statistics textbook. Determine if the raw score is above or below the mean Z-scores below the mean are expressed as negatives. ... Find the z-score value on the chart Z-scores are expressed through the tenths decimal on the vertical axis. ... More items...
https://www.usablestats.com/lessons/invz
The question doesn't state whether she wants at least the top 30% or at max the top 30%, but the former seems reasonable. Choosing 0.53 as the z-value, would mean we 'only' test 29.81% of the students. I would have assumed it would make more sense to choose z=0.52 for that reason, so that we at least cover 30%.
https://www.math.uni.edu/~campbell/mdm/quart.html
doc for "z score of 90th percentile".(Page 1 of about 16 results)