Karteikarten
Research Skills data
Karten in diesem Stapel (44)
- 1DistributionsDistributions show us the frequency of each value in a sample
Distributions are important for inferential statistics
There are known distributions - 2Normal distributionA mathematically defined, theoretical distribution with a particular bell shape (Gaussian distribution)
Distribution is symmetrical on its middle axis
Mean, median and mode will be the same
Most interval measures should produce normally distributed data
Height, weight, reaction time etc.
But also standardised test scores and IQ etc. - 3SkewnessSkewness is the measure of asymmetry in distributions
If a data set is skewed, the mean will be further in the direction of the skew than the median which will be further in the direction of the skew than the mode.
If a data set is skewed, the median line in the boxplot will not be in the middle of the box - 4Positive skewmost values are clustered at the low end of the scale
- 5Negative skewmost values are clustered at the high end of the scale
- 6KurtosisKurtosis refers to the peak and dispersion of the data; how “squished” it is
- 7Leptokurtic datathe distribution is more peaked
- 8Platykurtic data:the distribution is more dispersed
- 9ProbabilityProbability: the likelihood of the occurrence of an event or outcome
p = number of ways the event could arise
/number of possible outcomes - 10In statistics, probability runs fromIn statistics, probability runs from 0 to 1 and is expressed in decimals
1 indicates a 100% probability of an event
0.5 indicates a 50% probability of an event.
0.05 indicates a 5% probability of an event. - 11The things we measure in psychology are empirical probabilityThere are too many variables to account for
Empirical probability, also known as experimental probability, refers to a probability that is based on historical data. In other words, empirical probability illustrates the likelihood of an event occurring based on historical data. - 12logical probabilityProbability is a measure of inductive support or partial entailment, generalizing both deductive logic's notion of entailment and the classical interpretation's assignments to “equally possible” cases. In Carnap's notation, c(h,e) is a measure of the degree of support that evidence e confers on h.
- 13Normal distribution68% of scores lie between -1 SD and +1 SD (68.26% to be precise (34.13%+34.13%)
95% of scores lie between -1.96 SD and +1.96 SD.
If you know the mean and SD of normally distributed data, you can work out how many data points there should be for any score - 14Sample sizeWe can use the properties of the normal distribution to simulate datasets with the same mean and SD as our Psych 1B sample
- 15Probability calculationsWe can also calculate the probability of a certain value in a perfectly normally distributed dataset
pnorm(q = 157.48, mean = 168.4, sd = 9.31, lower.tail = FALSE) - 16Why learn data skillsPart of BPS accredited curriculum
Necessary to understand, evaluate and conduct psychological research - 17Descriptive statisticsSummarise and describe a given dataset
Central tendency and spread
Concrete and known values - 18Inferential statisticsUses probability to infer/draw conclusions about a larger population from a smaller sample of data
Abstract/estimated/inferred values - 19Levels of measurementHow much detail do the data convey – some levels of measurement give you more information than others
It’s important to think about this before you design your experiment/measures
Nominal
Ordinal
Interval
Ratio - 20Nominal dataAlso known as categorical data
The lowest level of information: your data are split into categories, but we don’t know anything about the numerical relationships between those categories.
Only tells us the datapoint is in one category: groups are separated - 21Ordinal dataRanked positions in a group; numbers do not represent quantities or counts
We don’t know anything about the interval (or distance) between those positions (winner/second/third) - 22Interval dataInterval data use equal units: the distance of scale points is the same across the whole scale
Can calculate a range of summary statistics (mean, median etc.) - 23Ratio dataInterval data that has a true zero and can be used in ratios
If it can take a negative value, it isn’t ratio
Can be halfed - 24The great Likert scale debateScale/questionnaire data can be ordinal or interval – it depends on the measure
Single-item Likert scores should be treated as ordinal
Multiple Likert scores that are combined may be treated as interval
There is much disagreement about this! - 25Discrete dataCan only have a fixed number of values
Generally(Normal/Ordinal) - 26Continuous dataTheoretically can have an infinite number of values
Generally(Interval/Ratio) - 27What do descriptive statistics describe?You need to look at and report both to get a comprehensive picture of the data
Central tendency
Dispersion - 28Spread/dispersionHow much do the values vary around the central value
Range
Interquartile range (IQR)
Variance
Standard deviation - 29Central tendencyWhat is the most typical/representative score?
Mean/Median/Mode - 30Mean- The arithmetic average
- The sum of the scores divided by the number of scores.
Mean is the most commonly used and most sensitive measure: its value is directly affected by the magnitude of each score
However, this sensitivity can be a problem if you have outliers
? ̅=(∑?)/? - 31Median- The middle value: a point that divides a set of scores into equal halves
-Half the scores are above the median and half are below it
-If an even number of data points, take average of two middle numbers
If your data contains extreme values, we use the median because the mean can be heavily influenced by such outliers - 32Mode-The score that occurs most frequently.
-This is the crudest of all measures of central tendency. - 33Central tendency and distributionMean, median, mode and distributions are all related.
If a data set is symmetrical, then the mean, mode and median are all the same - 34Range- The difference between your highest and lowest score
- Very sensitive to outliers - 35Interquartile range (IQR)-Data that is between the 25th and 75th percentiles.
- Imagine ordering all the scores, dividing it into quarters and then chopping off the extreme upper and lower quarters. - 36Variance- Uses information from all the scores in a data set
- s^2 - the average squared deviation from the mean
?^2=(∑(?−?- )^2 )/(?−1) - 37Standard deviation- Square root of the variance
- Shows the average deviation from the mean puts variance back into the original scale/values measured
?x=((∑(?−?- )^2 )/(?−1)) - 38BoxplotsThe five-number summary:
Lowest value: Bottom of whisker
Lower Quartile (LQ): Bottom of green box (lowest 25% of data)
Median: Thick black line
Upper Quartile (UQ): Top of green box (upper 25% of data)
Highest value: Top of whisker
IQR = Length of green box - 39Levels of measurementDescriptive statistics change depending on the level of measurement
Nominal data = frequency
Ordinal data = frequency, mode, median
Interval/ratio = mean, SD - 40∑Sum
- 41?-Mean
- 42NCount
- 43S^2Variance
- 44S?Standard deviation
Informationen
- Beschreibung
- Year 1 Psychology-1B Karteikarten am Research Skills data, erstellt von Isaac Gilbert am 21/02/2022.
- Erstellt von
- Isaac Gilbert
- Aufrufe
- 0
- Bewertung