Research methods booklet three Flashcards
What is nominal data?
A simple frequency count for distinct categories when something can only belong to one category
What is ordinal data?
Where numbers can be placed in ascending or descending rank order. We do not know the true intervals between each position.
What is interval data?
Measurements are taken from a scale where each unit is the same size and the gap between each unit is fixed and equal. This is the most sophisticated type of data
What is a strength of nominal data?
Mode average.
Easy to generate from closed questions and large amounts of data can be collected quickly.
What is a weakness of nominal data?
Because it does not give a numerical score for each pt, this crude type of data does not permit sensitive analysis
What is a strength of ordinal data?
Mode + Median average
Indicates relative values on a linear scale instead of just totals, so it is more informative than nominal data
What is a weakness of ordinal data?
Subjective data as scaled data is open to interpretation. For example, on a scale of 1 - 5 of attractiveness, 4 could mean different things to different people
What is a strength of interval data?
Mode, median and mean average
More informative as points are directly comparable because they are all of equal value.
Scientific measures used to record the distance between values are highly reliable
What is a weakness of interval data?
Interval scales that do not contain scientific measurements means that there is no absolute baseline to the scale, so scoring 0 doesn’t means pts don’t possess the variable, just that the scale doesn’t measure it
Why do we have to complete statistical testing?
Researchers use statistical tests to determine the likelihood that the difference or relationship they have found has occurred due to chance
Why do we use the significance level of p _< 0.05?
We are willing to accept the 1 in 20 likelihood that the results could be down to chance factors. We are 95% confident that we have a true relationship or difference
What is a type I error?
When we wrongfully accept the experimental hypothesis. We believe that there is a difference or relationship when actually no such relationship exists. This is known as a false positive or an error of optimism. This occurs when the probability level is too wide.
What is a type II error?
When we wrongfully accept the null hypothesis, when in fact a relationship does exist. This is known as a false negative and occurs when the probability level was too narrow.
How does the chosen significance level cause errors?
p<0.05 strikes an acceptable balance between a type 1 or type 2 error. The stricter the significance level the more likely it is to make a type 2 error. The less stringent the more likely we are to make a type 1 error.
What is reliability?
How well research can be replicated at another time. It is the consistency of the research findings.
What is external reliability?
Producing the same results each time the test is carried out both over time and from each individual who takes it
What is internal reliability?
This is concerned with the consistency within a test. It is usually associated with questionnaires and tests, ensuring different questions are all measuring the same thing
What is observer reliability?
When 2 or more observers produce the same record of their observations
What is the split half method to assess reliability?
Compares a person’s performance on two halves of a test or questionnaire. You would expect a person’s performance on the first half of the test to be consistent with their performance on the second half of the test. This assesses internal reliability.
What is the test retest method of assessing reliability?
Where the test is repeated several times using the same pts. you would expect the same or very similar results each time from the same individuals. You would correlate the results and obtain a strong positive correlation. This assesses external reliability.
What is inter rater reliability?
Where the observations of two or more observers are compared for similarity. The observers would be trained in the use of a coding system so they can identify exactly what they are looking for. You would hope to obtain a strong positive correlation. This assesses observer reliability.
What is correlational analysis?
This could be a relationship between two halves of a test, two observer’s observations, the results of a test when repeated again. If there is reliability then we would expect a high correlation of at least 0.8.
How can you improve reliability?
Take multiple measurements - making a mean average reduces the impact of anomalous scores
Use of standardised procedures - this would allow for more careful replication which will verify whether the findings are consistent
Train researchers carefully - Make sure you specify what counts as key behaviours, video some and train researchers carefully to create consistency.
What is validity?
The extent to which research is measuring what it intends to measure
What is internal validity?
The extent to which our findings are due to the manipulation of the IV and not any other uncontrolled variables
What is temporal validity?
A type of external validity that concerns the extent to which research findings hold true over time
What is ecological validity?
The extent to which an experimental effect can be generalised from the study to other settings and situations
What is population validity?
The extent to which findings can be generalised from the sample tested to other populations
What is face validity?
The extent to which research looks as though it is doing what it claims to on the surface level
What is concurrent validity?
The extent to which a new measure compares to a previously validated measure