Paper 1 - Data/statistics Flashcards
what are measures of central tendency ?
averages of the data
mean, median and mode.
what are the advantages and disadvantages of mean ?
- it is affected by outliers.
- it uses every value in the data, so represents the data well.
what are the advantages and disadvantages of median ?
- less effected by outliers.
- does not account for the distance between values.
what are the advantages and disadvantages and mode ?
- can be used for nominal and ordinal data.
- when there is more than one mode it does not reflect the centre of data well.
what are measures of dispersion ?
the spread of data
range, variance and standard deviation
what are the advantages and disadvantages of range ?
- easy to calculate
- effected by extreme values
what is variance ?
measurement of the spread between numbers in a data set.
standard deviation squared
what is standard deviation
the square root of variance
how dispersed the data is from the mean
what are the advantages and disadvantages of variance and standard deviation
- all values are taken into account
- complex to calculate
what do bar charts represent ?
nominal data
what do histograms represent ?
continuous data
what to line graphs represent ?
continuous data
what do pie charts represent ?
nominal data
what do scatter graphs represent
correlational data
what does ∝ mean
proportional to
what are inferential statistics ?
procedures for drawing logical conclusions about the target population that the sample was drawn from.
what’s the observed value ?
the value produces after applying an inferential test formula.
what’s probability ?
the measure of the likelihood that an event may occur.
what is the significance level ?
the level of probability at which it has been agreed to reject the null hypothesis. 0.05.
what’s a type one error ?
when the researcher rejects the null hypothesis that is true.
what’s a type two error ?
when the researcher accepts a null hypothesis which is not true.
What’s quantitative data and its advantages/disadvantages ?
Data in numbers.
-Its easier to analyse because data is numerical which can be summarised using descriptive statistics (averages and graphs). This makes it easier to draw conclusions from the study as the data is represented clearly.
-It oversimplifies reality and human experience as it suggests that there are simple answers.
What’s qualitative data and its advantages/disadvantages ?
Non numerical data, e.g. pictures or words.
-Represents the true complexities of human behaviour because they’re not reduces to numbers.
-Provides rich detail of how people think and behave. This increases validity.
-More difficult to detect patterns and draw conclusions from because of the large variety of information collected.
What are the 3 levels of data and what do they mean ?
nominal- categories
ordinal - ranked
interval - public units of measurements
what does a small standard deviation mean ?
scores are close to the mean
what does a large standard deviation mean ?
scored are far from the mean - lots of individual differences
What are the steps for the wilcoxon test ?
- calculate the difference between each pair of scores. if the difference is zero ignore it in ranking.
- rank the differences from low to high ignoring negatives.
- add up the ranks for positive differences. then for the negative differences.
- chose the smallest value (observed value).
- compare it to the critical value table and write the significance statement.
in the wilcoxon test, does the observed value need to be lower or higher than the critical value ?
equal to or lower than.
what data is used to calculate the chi squared test
nominal
what is a positive skew
most of the scores are towards the left (lower than the mean)
what is a negative skew
most of the scores are on the right (higher than the mean)
how would you make estimates from data
- round the numbers
- do the calculation
how do you make order of magnitude calculations
- work out the apporximate value for all numbers (rounding the number)
- see how many times bigger or smaller the number is
what are non parametric tests
- inferential tests like binomial sign test, chi squared, wilcoxon, mann whitney U and spearmans Rho
what are the steps for the Binomial sign test
- put a + if the number in the column A is bigger then the number in column B
- put a - if the number in column A is smaller.
- add up the pluses and minuses and select the smallest value for the observed value s
what is the symbol for the observed value of the binomial sign test
s
for the binomial sign test what must the observed value of s be for a significant result
s value must be equal to or less than the critical value
when would you use the binomial sign test
- nominal, ordinal, interval data
- repeated measures
- difference
when would you use the chi squared test
- nominal
- independent measures
- correlation
when would you use the wilcoxon test
- ordinal, interval
- repeated measures
- difference
when would you use the mann whitney u test
- ordinal, interval
- independent measures
- difference
when would you use the spearmans rho test
- ordinal, interval
- independent
- correlation
what are the steps of the mann whitney u test
- rank each column
- add up the ranks for each column
- choose the smallest value
- calculate the smallest U equation which will be given in the exam (N1 = number of things in first column, N2 = number of things in second column)
what is the symbol for the observed value in the mann whitney u test
u
in the mann whitney u test when is the observed value of U significant
u value must be equal to or less than the critical value
what are the steps for the Chi squared test
- calculate the totals for each row and column
- find the observed value of x squared using the equation: sum of (E-O)squared divided by E
- calculate the degrees of freedom for rows and columns and multiply it
for the chi squared test observed value to be significant what are the conditions
- observed value must be equal to or greater than the critical value
what are the steps for the spearman’s rho test
- rank each column
- calculate the difference between the ranks
- square this
- add up the total
- find the observed value of rho using 1-6 multiplied by the total divided by N(Nsquared -1)
for the observed value of the spearmans rho test to be significant what are the conditions
observed value must be greater than or equal to the critical value