understanding research

0.0(0)
Studied by 0 people
call kaiCall Kai
learnLearn
examPractice Test
spaced repetitionSpaced Repetition
heart puzzleMatch
flashcardsFlashcards
GameKnowt Play
Card Sorting

1/6

encourage image

There's no tags or description

Looks like no tags are added yet.

7 Terms

1
New cards

qualitative vs quantitative

qualitative data is a summary of verbal findings (words)

  • deals with experiences and perceptions

  • difficult to standardize and compare

  • some concerns about bias affecting findings

quantitative data is based on numbers and statistics

  • fewer issues with bias

  • reliable when done well


2
New cards

reliability + 2 types

  • inter-rater

  • test-retest


how consistent the result of a test will be; 2 types

  • inter-rater

    • imagine judges at the Olympics

    • 6, 7, 6 == high inter-rater reliability, 0.9

    • 10, 6, 3 == low inter-rater reliability, 0.2

  • test-retest

    • 1 judge, scoring the same person multiple times

    • 6 (2 weeks), 6 (2 weeks), 7 == high test-retest reliability, 0.8

    • 5 (2 weeks), 2 (2 weeks), 9 == low test-retest reliability, 0.3


3
New cards

validity + 2 types:

  • concurrent validity

  • predictive validity


how well does a test measure what we want it to measure

  • concurrent validity

    • does this compare well with other measures of the same thing?; does this concur with other measures of the same thing?

    • e.g., scores compared with another motor skills assessment?

  • predictive validity

    • does this predict actual performance?

    • e.g., ADL assessment compared to actual assistance required


4
New cards

correlation

measure of how well something predicts something else

  • there are positive and negative correlations


<p>measure of how well something predicts something else </p><ul><li><p>there are positive and negative correlations </p></li></ul><p></p>
5
New cards

standard deviation

2 SDs or more below the mean= services are warranted

<p>2 SDs or more below the mean= services are warranted </p>
6
New cards

p-value

tells us how reliable our data is; technically, it’s the probability that the data we have is just a product of change

  • a low p-value (less than 0.05) means a low probability our results happened by chance (less than 5%)

  • a high p-value (greater than 0.05) means a high probability our results happened by change (greater than 5%)


<p>tells us how reliable our data is; technically, it’s the probability that the data we have is just a product of change </p><ul><li><p><strong>a low p-value (less than 0.05) </strong>means a low probability our results happened by chance (<strong>less than 5%</strong>)</p></li><li><p><strong>a high p-value (greater than 0.05) </strong>means a high probability our results happened by change (<strong>greater than 5%) </strong></p></li></ul><p></p>
7
New cards

percentile rank

a way to rank something, for example, the motor skills of a 6-year-old child

  • e.g., a child in the 7th percentile means that

    • 7% of children their age score worse

    • 93% of children their age score better

  • just remember:

    • high percentile is a high rank

    • low percentile is a low rank