Uncovering the Truth: The Difference Between Random and Systematic Errors

[Khawla]

A human error, is it random or systematic error?

 

[nicksauce]

I would say neither. Random errors are natural errors. Systematic errors are due to imprecision or problems with instruments. Human error means you screwed something up, you made a mistake. In a well-designed experiment performed by a competent experimenter, you should not make any mistakes.

Although maybe you can give a specific example of "human error", as the term could be ambiguous.

 

[rock.freak667]

That's a random error I believe.

EDIT: I assumed by "human error" you meant things like judgements like when looking for a color change in a titration.

 

[Khawla]

Nicksauce,
we only had two options in the exams, random and systematic. So it's either random or systematic.

Anyways, after I checked our book I couldn't find any specific answer.
What I found is that the random errors might occur by the OBSERVER?
Does that mean that a random error is the correct answer?

My mother tongue language isn't English, that's why am asking :)

 

[nicksauce]

If "human error" is interpreted the way rock.freak667 says, then I would agree it's random error.

 

[D H]

A human error, is it random or systematic error?

My initial response: what a dumb question. Upon seeing that this is a question on an exam, I modified my response to "what an incredibly dumb question!".

Random errors have a mean of zero. Accumulate enough statistics and the random errors average out. Systematic errors do not have a mean of zero. Accumulating statistics will remove systematic errors.

Suppose you paint pairs of circles on a blank piece of paper, hand out some the papers and rulers to a group of people, and ask them to "measure the distance between the circles". Suppose that what you really want is the center-to-center distance. In asking for the distance between circles you opened yourself up to systematic error. A few people will measure from the outside edge of one circle to outside edge of another, resulting in consistently high answers. Some people read your mind and attempt to measure center-to-center. Even more people will measure inside edge to inside edge, yielding consistently low answers. The mean result will be low because you have introduced systematic error in the way you phrased the question. Suppose instead that you specifically ask people to try to estimate the center-to-center distance. This experiment will exhibit random errors because people will have some difficulty estimating where the centers lie.

Bottom line: people make both random and systematic errors.

 

[Khawla]

D H,
I tottaly agree! I think this question is dumb too!
Okay let's see, we study Kane and Sternheim Physics book, 3rd addition.
I looked for an answer in the first chapter but I couldn't find one.
But let me show you what's written in our book:

1- The variation in results about this average arises from the inability of the -observer- to start and stop the watch exactly the same way each time. the error introduced by this inability is -random- and can be reduced by taking the average of many measurments.

2- A -systematic- error may also result from the -observer-'s reaction time. the observer may systematically start or stop the watch too early or too late.

 

[HallsofIvy]

From which it follows that human error may be either random or systematic. Could you answer "both"?

 

[Khawla]

Actually no, the only two answers were systematic and random. There was no other choices in the question.

 

[Borek]

This qualifies as the "question that should be never asked". However,

Systematic errors do not have a mean of zero. Accumulating statistics will remove systematic errors.

Did you mean "will not remove"?

 

[Khawla]

Hey guys!
I forgot to tell you that I passed my exam afterall. And the correct answer (or what the Doctor think is corret) is Random ;p.