0

I have the following sentence:

Researchers have noted the predictive power of features derived from students' activity sequences when predicting if students will stop participating in online courses.

I know that predicting sounds a little repetitive. However, I could not find another way to phrase this. Any ideas?

  • I'm not clear what is meant by predicting engagement. Do you mean ...when engaged in online courses? (The phrase Interaction sequences is also somewhat puzzling!) – Ronald Sole Mar 31 '17 at 15:00
  • Sorry. These are domain terminologies. I updated it, does it look better? – renakre Mar 31 '17 at 15:02
  • 1
    You could replace predicting if with assessing/forecasting/evaluating whether. – Ronald Sole Mar 31 '17 at 15:07
  • 1
    Consider predictive power ... regarding when students .... – Lawrence Mar 31 '17 at 15:13
  • 1
    What about "...predictive features...used to determine whether students will stop..."? – Michael R Atchley Apr 1 '17 at 15:25
1

The current structure is

  • the predictive power of (data) when (situation).

Since the situation is predicting, you end up using predict twice, once to specify the type of power, and then to elaborate on its application. The problem is that the elaboration also specifies the type of power, leading to the redundancy you've pointed out.

To use predict just once in that sentence, drop one of the 'specifications':

  1. the power of (data) to predict (events); or
  2. the predictive power of (data) regarding (events).

The problem with #2 is ambiguity: the word regarding can refer to predictive power or to the data. You can try to overcome this by adding a comma before regarding, producing:

  • Researchers have noted the predictive power of features derived from students' activity sequences, regarding when students will stop participating in online courses.

I prefer #1 if the subtle change to the flavour of the sentence, from "the predictive power of features" to "the power of features to predict", is acceptable:

  • Researchers have noted the power of features derived from students' activity sequences to predict when students will stop participating in online courses.

If assigning power to features seems a little out of place in an academic work, consider editing the sentence further. It might also be useful to make the features more prominent in the sentence, relegating the researchers to a reference:

  • Features derived from students' activity sequences can be used to predict when students will stop participating in online courses (Rhea Searcher 2016).
  • @renakre Glad to help. :) – Lawrence Mar 31 '17 at 15:49
  • I wonder if power sounds a little strange there. Would capacity or ability be more appropriate? Or, it would not matter? What do you think? – renakre Mar 31 '17 at 15:51
  • @renakre Ah, that was the subtlety I mentioned. It's a bit stark to call it a power, isn't it? The question is how you wish to deal with the features - are they used or do they have capabilities? If the former, consider: "Researchers have used features ... to predict ...", or (if the specified researchers didn't actually use those features but just noted their potential) "Researchers have noted that features ... can be used to predict ...". This is part of the joy of editing - you end up saying what you really mean. – Lawrence Mar 31 '17 at 15:54
  • Oops, sorry - I edited my comment while you were making another. I think the edited comment addresses your last query. – Lawrence Mar 31 '17 at 15:57
  • Haha, that is right! :) How about: "Researchers have found that features... can be powerful predictors of ...". (see it all changed, but I guess this is what I really mean, as you said :) – renakre Mar 31 '17 at 16:00

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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