alan bradley [email protected] alan.bardely researching society and culture week 5...
TRANSCRIPT
Alan Bradley
Alan.Bardely
Researching society and culture
Week 5 Quantitative methods (2)
Lecture Outline
• A brief recap of some terms from last week
• Types of quantitative research
• Surveys and questionnaires
• Samples and pilot studies
• Measuring concepts and indicators
• Analysing quantitative data
Some terms from last week
• 1) Representativeness – relates to sample• 2) Replication• 3) Validity • Internal:- Are causal relationships valid?• External:- are results generalisable?• 4) Reliability – consistency of measures of
concept• More on all of this as we go through the
lecture
Types of quantitative research
• 1) Primary data – examples are surveys and questionnaires, structured interviews, content analysis.
• 2) Secondary data – a) collected by other researchers or: b) collected by institutions.
• The focus in this lecture will be on surveys/questionnaires
The process in surveys/questionnaires
• 1) Theory and hypothesis• 2) Research question(s)• 3) Research design – methods etc• 4) Choosing sample – issues raised last week• 5) Questionnaire – open/closed question, self completed,
mail or face to face, interviewers• 6) Pilot study• 7) Data collection• 8) Data analysis • All of this is an ‘ideal type’
Sampling and Pilot study
• Sample aims to represent the whole population (remember last week)
• Pilot study to check that questions can be understood, no ambiguity, tests the design of the research.
Measuring concepts
• Some concepts relatively easy to measure, eg sex/gender?
• Others may be more problematic, eg social class, social mobility, poverty
• ‘Measurement (of concepts) provides the basis for more precise estimates of the degree of relationship between concepts’
• (Bryman, A. 2008: p144)
Indicators of concepts
• Measures are quantities (eg household income)
• Indicators are concepts which allow us to quantify non-quantifiable concepts
• That is, indicators are used as if they are concepts.
• What indicators may allow us to measure social class, poverty, attitudes?
Indicators of social class
• Employment status commonly used:• 1) Higher management• 2) Lower management• 3) Intermediate occupations• 4) Small employers/self-employed• 5) Lower supervisory/lower technical occupations• 6) Semi routine occupations • 7) Routine occupations• National Statistics Socio-economic Classification• Think what other indicators could be used.
Analysing quantitative data
• Univariate - frequency tables/central tendency. Example to follow.
• Bivariate – eg gender and educational attainment
• Multivariate – computer analysis needed
• 2 and 3 raise issues of correlation versus cause and effect
Univariate analysis
• Usually frequency tables showing the number of instances occurring, and the % of the whole.
• May include grouped variables:-• Age 20 and under
21-30
31-40
41-50
51 and over Why these intervals?
Univariate analysis (2)
• Measures of central tendency (Averages)
• Mean - total divided by number in group
• Mode - figure that appears most often
• Median – in the middle, ie the same number above as below
Example of central tendency
• Annual income in £ 1) 5,000
2) 7,000
3) 20,000
4) 20,000
5) 30,000
6) 50,000
7) 200,000• Calculate mean, mode and median incomes
Answers
• Mean = 47,430
• Mode = 20,000
• Median = 20,000
• How useful are these?
Measures of dispersion
• Range in previous example is:-
• 5,000 to 200,000
• Thus, many of the instances vary widely form the average.
• Standard deviation measures the average amount of variation from the mean.
• Issues of validity etc
• Calculations carried out by computer (SPSS)
Suggested readings
• Bryman, A. 2008 Chapter 8 (The nature of quantitative research)
• and Chapter 14, just the bit on uni/bivariate analysis
• OR:- search in any methods textbook for chapters on quantitative data collection and analysis!
Next week
• Qualitative research - interviews