Measures of central tendency
A single cental score thay enables a ummarise the distribtuon of a data set. Describe the centre of a distribution of scores: these ar the mode, median and mean
Measures of variability
Describes the extent to which individuals or scores in the sample vary. The most common measures used are the range, variance, standard deviation, and semi-quartile range
Inferential stats
Enable inferences and conclusions to be drawn from those data to estimate the probability that statistics found in a random sample accurately reflect the population parameter; and to test hypotheses about a population
Descriptive stats
Procedures that allow researchers to describe, organise and summarise raw data
Levels of measurement
Measurement level is determined by the nature of the object or event being measured. Levels of measurement in ascending order are; nominal, ordinal, interval, and ration (acronym NOIR)
Nominal measurement
When data are at the nominal level and the researcher wants to determine whether groups are proportionally different, the chi-square x2 statistic is used
Ordinal measurement
Used to show the relative ranking of objects or events in a hierarchical order. Numbers assigned to each category can be compared and a member of a high category can be said to have more of a attribute than one in a lower category
Interval measurement
The differences between scores or measures can be treated as equal, as there is specific numerical distance between each of the levels. However, the zero point remains arbitrary, reflecting the freezing point of water which is not he same as having no hea
Ratio measurement
Shows ranking of events or objects on scales with equal intervals and absolute zeros; the zero point makes the ratio of the scale values meaningful. Highest level of measurement, usually only achieved in the physical sciences
Normal distribution
The concept of the normal distribution is a theoretical one. It is based on the observation that data from repeated measures of some interval or ratio level data group themselves about a midpoint in a distribution in a manner that closely approximately th
P value
Significance is reported using p value and gives the probability that an beseeched difference happened by chance alone
Confidence intervals
To help the researcher assess what the values might be in the target population. Provide plausible range of values bracketed by lower and upper limits that encompass the unknown population or true mean estimated by that sample mean.
Standard deviation
Most frequently used measure of variability and it is based on the concept of the normal curve. It is a measure of standard deviation or distance of each score from the group mean in a normal distribution and as such must be always reported by the mean
T test
2 types. The independent group t-test tests for difference in mean outcome between two mutually exclusive group e.g. males or females. At one time point only.
A paired t test tests for difference between the one group at two separate time points e.g. a pr
Inclusion/exclusion
Define specific characteristics of participants which determines inclusion/exclusion into the study sample.
Random selection
How study participants are selected from the population for inclusion in the study. Aims to ensure that each participant in the population has an equal chance of being selected in the sample
Randomisation assignment
Designed to ensure that groups or participants are similar on the variables of interest, so that difference in the outcome variable can be attributed to the intervention
Randomised control trial
Clinical equivalent of a true experiment, the gold standard for testing cause and effect relationships in clickable research
Generalisability (external validity)
A researchers findings may be entirely valid in one setting but not in another. Describes the content to which research findings can be applied to settings other than that in which they were originally tested
independent variable
What the researcher or nature manipulates, a treatment, program Or cause
Dependant variable
What is effected by the independent variable - your effect on outcomes
null hypothesis states ..
That there is no relationship between the independent and dependant variables
Probability sampling ensures..
That all elements have an equal chance of being seated in the sample, therefore minimising bias and enabling generalisability of the findings