Yanyan is designing a performance evaluation tool for staff at her research institute. She wants to capture the quantity and quality of their research output, as well as their success in funding applications. She has access to the following data:

1) High quality publications - papers published by each researcher in respected peer review journals. (ranges from 2 to 15)

2) Low quality publications - papers published by each research in journals with poor standing or that have no peer review process. (ranges from 0 to 21)

3) Funding - the amount of external funding the researcher has been able to obtain. (ranges from $0 to $25000)

Which of the following are appropriate data transformations that could be used? [Multiple answers could be selected]

Question 3 options:

	A decreasing function, N(t) = 21 - t, can be used for the low quality publications variable, after which it could be scaled so that it ranges from 0 to 1.
	Linear feature scaling (without any negation) can be used on the funding variable, i.e. of the form {"version":"1.1","math":"$f\left(t\right)=\frac{t}{25000}$"}, so that the data ranges from 0 to 1.
	Linear feature scaling (without any negation) can be used on the high quality publications variable, i.e. of the form {"version":"1.1","math":"$f\left(t\right)=\frac{t-2}{15-2}$"}, so that the data ranges from 0 to 1.
	A decreasing function, N(t) = 15 - t, can be used for the high quality publications variable, after which it could be scaled so that it ranges from 0 to 1.
	A decreasing function, N(t) = 25000 - t, can be used for the funding variable, after which it could be scaled so that it ranges from 0 to 1.
	Linear feature scaling (without any negation) can be used on the low quality publications variable, i.e. of the form {"version":"1.1","math":"$f\left(t\right)=\frac{t}{21}$"}, so that the data ranges from 0 to 1.