0
$\begingroup$

I'm doing an experiment investigating the Mpemba effect. I measured some factors that may have contributed to the time it took the water to freeze (defined as the point where it becomes a total solid, i.e. at the end of the phase change), such as the time it took to reach 0 degrees Celsius, the lowest temperature it reached before freezing (i.e. how much it supercooled), and the time it spent at the freezing temperature.

I'm not sure what to call those variables since I couldn't vary them, which means it's not an independent variable. I'm not sure if they are dependent on the initial temperature because there are still so many things unknown about the Mpemba effect and factors which I couldn't control such as the roughness of the container that it was in (which contributes to the supercooling). And they're obviously not controlled variables since I couldn't control them.

But my best guess is that it barely falls under either independent or dependent variable.

$\endgroup$
3
  • $\begingroup$ Are you suggesting that time is not an independent variable? $\endgroup$
    – joseph h
    Jul 31, 2021 at 3:14
  • $\begingroup$ Time is a dependent variable, I'm not sure whether the rest are $\endgroup$
    – prata
    Jul 31, 2021 at 3:49
  • 1
    $\begingroup$ Ah. I see what you are doing. You have made time the dependent variable (how long it takes the water to freeze) and are investigating what other possible variables can affect it. Nice. $\endgroup$
    – joseph h
    Jul 31, 2021 at 3:54

2 Answers 2

0
$\begingroup$

What you call them is not important. What is important is how you use them, what statistical methods you use and so forth.

The most important thing to consider is how accurately they are known. In a standard linear model or least squares regression, the explanatory variables are assumed to have no uncertainty. So if you plan on using such techniques then they should be things that you can know far more precisely than your predicted variable.

If that is not the case then you may need to use other methods, such as error in variables techniques.

$\endgroup$
1
  • $\begingroup$ I would use a better method, but the causes of the Mpemba effect are still unknown. On top of that, I did this in a high school laboratory, not much apparatus which I could use to make it better. $\endgroup$
    – prata
    Jul 31, 2021 at 3:07
0
$\begingroup$

Here are two suggestions - if you read about each, you can decide which bests fits your situation.

  1. nuisance variable / nuisance parameter
  2. extraneous variable

Hope this helps.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.