comparison quantitative and qualitative
DESCRIPTION
Research methodsTRANSCRIPT
Quantitative Research Methods Training
Comparing Qualitative and Quantitative Methods
Edema Conference Centre, Morogoro
4th August, 2014
What is research?• Systematic inquiry that investigates hypothesis• Suggests new interpretation of data/texts or
new knowledge – Poses new questions for future research to explore
• It also involves – Asking questions that nobody has ever asked before– Doing necessary work to find the answer– Communicating the knowledge you have acquired to
a larger audience
Methods 1) Quantitative
2) Qualitative
Quantitative methods Are driven by two concerns;• Nature of relationships among variables
– Unrelated – Correlation– If one variable affect another– Independent and dependent variable(there is effect)
• Whether discoveries or findings are generalizable– Study design– Sampling procedures– Analysis of data
Quantitative methods…• Would want to quantify the two concepts
– The size of relationships among variablesThis is quantified through mathematicso The difference average scores of maths between female and
maleo The correlation between scores on an IQ test and grade point
average
– The probability that the results are generalizable
This is quantified through inferential statisticso There is statistically significant difference at 0.5 level between
male and female on mathematics scores
Quantitative methods…What quantitative research answers
• Research question demanding a quantitative answer e.g how many, to what extent
• Numerical change(down or up)
• State of something e.g. factors affecting recruitment of CHWs?
• Testing of hypotheses e.g. understand relationship between variable A&B
Comparison
General framework Quantitative Qualitative
Seek to confirm hypotheses aboutphenomena
Seek to explore phenomena
Instruments use more rigid styleof eliciting and categorizingresponses to questions
Instruments use more flexible,iterative style of eliciting andcategorizing responses to questions
Use highly structured methodssuch as questionnaires, surveys,and structured observation
Use semi-structured methods suchas in-depth interviews, focusgroups, and participant observation
More generalizable Less generalizable
More objective: provide observed effects(interpreted by researcher)
More subjective: describe a problem from the respondents point of view
Analytical objectivesQuantitative Qualitative
To quantify variation To describe variation
To predict causal relationships To describe and explain relationships
To describe characteristics of apopulation
To describe individual experiences
To describe group norms
Question format Quantitative Qualitative
Closed-ended Open-ended
Fixed responses options Unstructured or semi-structured responses options
Data format Quantitative Qualitative
Numerical (obtained by assigningnumerical values to responses)
Textual (obtained from audiotapes,videotapes, and field notes)
Less in-depth but more breadth of information across a large number of cases
More in-depth information on a few cases
Flexibility in study designQuantitative Qualitative
Study design is stable frombeginning to end
Some aspects of the study areflexible (for example, the addition,exclusion, or wording of particularinterview questions)
Participant responses do notinfluence or determine how andwhich questions researchers asknext
Participant responses affect howand which questions researchersask next
Study design is subject tostatistical assumptions andconditions
Study design is iterative, that is,data collection and researchquestions are adjusted accordingto what is learned
Sampling Quantitative Qualitative
Probability sampling Non probability sampling
Data collection and analysisQuantitative Qualitative
Time expenditure heavier on planning phase and lighter on the analysis phase
Time expenditure lighter on planning end and heavier during analysis
Can be valid and reliable; largely depends on measurements device or instrument used
Can be valid and reliable; largely depends on skill and rigor of the researcher
Statistical tests are used for analysis No statistical tests