Flashcards
Quantitative Research
Cards in this deck (13)
- 1What is a model in a quantitative study?- a model is usually based on theory and helpful to generate hypotheses
- a model is a representation of the real world but rarely perfect
- it is a parsimonius representation of a phenomenon
- model is a close as possible to the data but there are some errors or residuals remaining
- goal is to minimize these errors and get the best prediction
Data = model + error - 2What is robustness in a quantitative study?- the extent to which a summary measure is sensitive to disturbances in data quality
- disturbances in data quality can arise either from small changes, too many data values or by large errors in a few data values
- mean is not robust, but the mid-mean and media are - 3What is efficiency in a quantitative study?- How much did the researcher get out of his data?
- efficiency refers to how much a study measure captures all the information within the data that is relevant to the study - 4What types of variables could you find in a quantitative study?1. categorical
a. nominal variables: variables that have two or more categories but which do not have an intrinsic order (types of industries)
b. ordinal: variables that have two or more categories, but the categories can be ordered or ranked (agree, disagree, average)
c. dichotomus: normal variables which only have two categories or levels (gender)
2. continuous
a. variables for which central characteristic is that they can be measured along a continuum + they have numerical value - 5When do we use a t-test in a quantitative study and what does it tell you?- used to compare groups
- compares the difference between the mean or average score of two independent groups on a given variable
assumptions
- the dependent variable is normally distributed
- the two groups have approximately equal variance
- the two groups are independent of one another
- determines if the means are sufficiently different from each other to say that they belong to two distinct groups
- - 6When do we use a correlation in a quantitative study and what does it tell you?- used to describe the observed relationship between instances of two events
- positive correlation means that as a variable increases, the other variable increases as well
- negative correlation means that as one variable stays, the other variable decreases - 7When do we use the analysis of variance (i.e. ANOVA) in a quantitative study and what does it tell you?- a form of hypothesis test for comparing the means of two or more groups which may be classified on the basis of other variables
- should only be used of the quality of measurement is high
- very efficient because it uses the mean - 8When do we use a regression in a quantitative study and what does it tell you?- is used to model the relationship between a dependent variable and an independent variable
- is used in linear relationships, where an increase/decrease of variable x results in an increase/decrease of variable y
- regression analysis determines the line of best fit (y=bx+c+e)
- relationship either positive or negative - 9What is a control variable?- in scientific experimentation it is the element which is constant and unchanged throughout the course of the investigation
- has a strong influence on experimental results
- is held constant to test the relative relationship between dependent and independent variable
- variable itself is not of primary interest - 10How do you ensure quality in quantitative research and why is it important?Researcher needs to take into account
a. measurement: How variables have been constructed?
b. causality: explaining causes and effects: independent variable, dependent variable, moderators, mediators
c. replication: explicitly state procedures so research can be done again
d. robustness: extent to which a summary measure is sensitive to disturbances in data quality
e. efficiency: extent to which a summary measure captures all the information within the data which is relevant to what is summarized
important to ensure the quality to make the research considered as useful and trustworthy - 11What are the quality criteria in quantitative research?validity
- integrity of conclusions derived from study internally and externally
reliability
- extent to which an instrument for measuring something gives consistent results (allows replication)
objectivity
- research is not influenced by e.g. personal bias, values and believes - 12What is a hypothesis?- a formal statement of the expected relationships among variables
- predicts a relationship among two or more variables
- it is testable
- it is justifiable (based on theory) - 13What are the formal steps in hypothesis testing?1. define a research hypothesis to be tested (reflecting purpose of the study)
2. define a null hypothesis (if primary hypothesis is inconclusive, this is the conclusion)
3. deriving a summary measure of a characteristic of interest
4. choosing a reference distribution and calculating a test statistic
5. drawing a conclusion (are the results significant to support hypothesis)
Information
- Description
- Flashcards on Quantitative Research , created by Peter Lustig on 16/12/2017.
- Created by
- Peter Lustig
- Views
- 4
- Rating