SYA4300 FINAL EXAM STUDY GUIDE Quantitative data analysis What is it Quantitative research involves the collection of data that is quantifiable countable mathematically calculated It creates a means for researchers to be able to generate statistics This type of research often generalizes data to a larger population of people generalizability Quantitative research allows us to provide an explanation for certain aspects of the social world stand for things we want to explain and give a basis for measuring variation Quantitative data analysis is a helpful way to evaluate and interpret results from quantitative research Before the analysis make sure to identify the level of measurement being used The three types of measurements are classified as nominal ordinal or interval ratio The quantitative statistics can be broken up into two different parts descriptive and inferential a Descriptive statistics describe the distribution of the population and relationship among the variables b Inferential statistics are used to estimate how likely it is that a statistical result based on data from a random sample is representative of the population Used to make predictions determine statistical significance Critique of quantitative research Feminist s critique says that it suppresses the voices of women and other marginalized groups people are seen as objects of research instead of subjects Univariate analysis frequency tables measures of central tendency mean median mode measures of dispersion range minimum maximum graphs pie charts bar charts Univariate analysis explores each variable in a data set separately Step 1 The data needs to be cleaned and organized Code identify missing numbers Step 2 Examine the distribution of your data We look at a Central tendency the value around the distribution of your data Mode most frequently occurring value can be nominal ordinal interval ratio will always be a whole Median median point within entire range of values can be ordinal interval ratio Mean sum of all values in distribution then divide by of values can be ordinal interval ratio thrown off by outliers b Variability the extent to which cases are spread out through the distribution or clustered around one value measures of dispersion Range the difference between the minimum and maximum values in a sample Interquartile range the range of values of a frequency distribution between the first and third quartiles Variance divergent inconsistent Standard deviation the average amount of variation around the mean reducing the impact of extreme values c Skewness extent to which cases are clustered more at one or the other end of the distribution of a quant variable rather than in a symmetric pattern around the center d Frequency tables of people or cases in each category usually expressed as of sample e Diagrams Bar chart or pie charts nominal ordinal Histogram interval ratio line graphs time trends Grouping values cleaning up data We group data to fit the type of graphical presentation ex More than 10 values are best for pie bar graph bar chart s are best fit for ordinal pie charts are best fit for nominal histogram s are best fit for interval ratio We also group data so the levels of measurement reflect the type of analysis Ex In cross tabs categories should not exceed 5 categories should be logically defensible categories should reflect concept categories should be mutually exclusive categories should be exhaustive try to make n about the same size in each group Bivariate analysis Cross tabulation contingency tables what is it making sense of the data interpretations identifying trends Step 3 In bivariate analyses we look at 2 variables in relation to one another This analysis searches for co variance correlation Bivariate analysis cannot establish causality but can sometimes infer the direction of causal relationships This analysis uses cross tabulations contingency tables Cross tabulations All about exploring the relationships These contingency tables are used as relationship indicators that connect the frequencies of two variables and helps to identify patterns of association PATTERNS OF ASSOCIATION 1 Existence Do the distributions vary at all between categories of the 2 Strength How much does the distributions vary between categories of independent variable Independent variables 3 Direction Do values on the dependent variable tend to increase or decrease w an increase in value on the independent variable 4 Pattern Are changes in the distribution of the dependent variable fairly regular increase or decrease or do they vary gradually decreasing then rapidly increasing Chi Square This test establishes how confident we can be that there is a relationship between the two variables in the population Pearson s r This is the strength of association in terms of interval ratio coefficients The value ranges from 0 1 determines a positive negative direction of association Multivariate analysis when to conduct This examines the relationship of multiple variables It is used for OLS regression logistic regression and ordered logistic Correlation and causation Interpreted through causality Correlation The strength of the relationship between two variables Causality The relationship occurs in a predictable way when one event leads to the other Spuriousness A false relationship between variables Mediating moderating intervening variables The mediator variable accounts for the relationship between the independent and dependent variable It identifies and specifies the process of an observed relationship between each variable The moderator variable affects the direction and or strength of the relation between the independent and dependent variable The intervening variables explains the internal relationship of the independent and dependent variables Ex X Y Missing data Incomplete responses missing values This occurs when no data value is stored for the variable in an observation Various Tests How to test statistical significance Tests Determines significance by setting up the null hypothesis the test decides if relationship does or does not exist Ex How risky is it to make this inference Decide on a level of acceptable statistical significance usually 0 05 Use a statistical test we used chi square for research paper to see if acceptable level is attained If level is attained reject the null meaning hypothesis is seen as repeatable relationship is statistically significant If level is not attained accept null hypothesis