PAC RESEARCH METHODOLOGY Quantitative Methods PAC 代写澳洲作业工业自动化课程论文 RESEARCH METHODOLOGY Quantitative Methods 1 INTRODUCTION RESEARCH AIMS & THE SURVEY METHOD Session Objectives To provide a brief overview of the key issues to be considered in survey work To introduce, define and illustrate some of the key terms used in quantitative research work To provide a basic theoretical underpinning for computerized quantitative data analysis to show the links between survey design, sampling and data analysis decisions SESSION STRUCTURE Survey design Data analysis Sampling … With plenty of exercises to keep you going…

 PAC RESEARCH METHODOLOGY Quantitative Methods 1 SURVEY DESIGN SURVEY PLANNING STAGES Assessing constraints & resources Structuring the problem Specifying population and sample Deciding on method of data collection Designing the questionnaire Pre-testing and piloting Carrying out the survey Data processing and analysis Writing the dissertation QUESTIONNAIRE DESIGN Defining theories and hypotheses in terms of likely relationships between variables Identifying data requirement & making a list of things to ask questions about Type of questionnaire & choosing the best alternative Question wording Question coding Presentation, order, lay-out and flow ADMINISTERING THE SURVEY IMPORTANT THINGS YOU SHOULDN’T FORGET: Introduction for self-administered questionnaires Covering letter for postal questionnaires Field worker training for interviewer administered and telephone surveys Follow-up activities What about delivery and collection questionnaires? Two key considerations for any survey (if you want to get a good mark)… Validity Reliability (Plus one more from my experience: …not finding the questionnaire too embarrassing when you look at it 10 years later…) PAC RESEARCH METHODOLOGY Quantitative Methods 1 DATA ANALYSIS … what’s a hypothesis anyway? PURPOSES OF DATA ANALYSIS To show the frequency of occurrence To examine the distribution of values To describe the central tendency and variability To compare variables To identify relationships between variables To test whether findings are statistically significant DATA ANALYSIS QUESTIONS What is a typical value? What is the spread of values? Is there a relationship between variables? How strong is the (linear) relationship? What are the chances that the findings are only an accident? ANALYSIS STRATEGY Work out measures of central tendency Work out measures of dispersion Plot data on a graph/do cross tabulations Examine correlation Do significance testing MEASUREMENT SCALES AND STATISTICS TYPE OF DATA SCALE Categorical Data Nominal Ordinal Quantifiable Data Interval and ratio PERMISSIBLE STATISTICS (e.g.) For Categorical Data: Frequency & mode… Plus median… For Quantifiable Data: Plus mean & standard deviation MEASUREMENT SCALES & SIGNIFICANCE TESTING TYPE OF DATA SCALE Categorical Data Nominal Ordinal Quantifiable Data Interval and ratio PERMISSIBLE TESTS (e.g.) For Categorical Data: Chi square For Quantifiable Data: T-test Correlation coefficient … it is vitally important to think about the way in which the data will be analyzed and presented at the design stage! PAC RESEARCH METHODOLOGY Quantitative Methods 1 SAMPLING CENSUS V. SAMPLE Cost Time Accuracy The destructive nature of measurement THE PROCESS OF SAMPLING Define the population Decide on the sample size Define a frame for that population Select a sampling unit Choose a method of selecting that sample Probability Non-probability Define the sampling plan Select the sample CHOICE OF SAMPLING METHOD PROBABILITY Simple random Stratified random Cluster Multi-stage NON-PROBABILITY Convenience Judgment Purposive Quota … the most important thing about your sampling decisions is that they determine the extent to which you can generalize from your findings. CONCLUSIONS In a survey you typically only get one shot. It is therefore important to allow plenty of time to think through all the potential issues of survey design, www.liuxuelw.com/aozhoulunwen/ data analysis and sampling at the outset. Quite a lot of further reading and hands-on practice may be needed… (That’s normal!) Additional resources General: Diamantopoulos, A. & Schlegelmilch, B.B. (1997) Taking the fear out of data analysis, The Dryden Press, London De Vaux, D. (2002) Surveys in Social Research, 5th edition, Rutledge, London Useful on questionnaire design: Google search: Learning spas

PAC RESEARCH METHODOLOGY Quantitative Methods PAC 代写澳洲作业工业自动化课程论文 RESEARCH METHODOLOGY Quantitative Methods 1 INTRODUCTION RESEARCH AIMS & THE SURVEY METHOD Session Objectives To provide a brief overview of the key issues to be considered in survey work To introduce, define and illustrate some of the key terms used in quantitative research work To provide a basic theoretical underpinning for computerized quantitative data analysis to show the links between survey design, sampling and data analysis decisions SESSION STRUCTURE Survey design Data analysis Sampling … With plenty of exercises to keep you going…

 PAC RESEARCH METHODOLOGY Quantitative Methods 1 SURVEY DESIGN SURVEY PLANNING STAGES Assessing constraints & resources Structuring the problem Specifying population and sample Deciding on method of data collection Designing the questionnaire Pre-testing and piloting Carrying out the survey Data processing and analysis Writing the dissertation QUESTIONNAIRE DESIGN Defining theories and hypotheses in terms of likely relationships between variables Identifying data requirement & making a list of things to ask questions about Type of questionnaire & choosing the best alternative Question wording Question coding Presentation, order, lay-out and flow ADMINISTERING THE SURVEY IMPORTANT THINGS YOU SHOULDN’T FORGET: Introduction for self-administered questionnaires Covering letter for postal questionnaires Field worker training for interviewer administered and telephone surveys Follow-up activities What about delivery and collection questionnaires? Two key considerations for any survey (if you want to get a good mark)… Validity Reliability (Plus one more from my experience: …not finding the questionnaire too embarrassing when you look at it 10 years later…) PAC RESEARCH METHODOLOGY Quantitative Methods 1 DATA ANALYSIS … what’s a hypothesis anyway? PURPOSES OF DATA ANALYSIS To show the frequency of occurrence To examine the distribution of values To describe the central tendency and variability To compare variables To identify relationships between variables To test whether findings are statistically significant DATA ANALYSIS QUESTIONS What is a typical value? What is the spread of values? Is there a relationship between variables? How strong is the (linear) relationship? What are the chances that the findings are only an accident? ANALYSIS STRATEGY Work out measures of central tendency Work out measures of dispersion Plot data on a graph/do cross tabulations Examine correlation Do significance testing MEASUREMENT SCALES AND STATISTICS TYPE OF DATA SCALE Categorical Data Nominal Ordinal Quantifiable Data Interval and ratio PERMISSIBLE STATISTICS (e.g.) For Categorical Data: Frequency & mode… Plus median… For Quantifiable Data: Plus mean & standard deviation MEASUREMENT SCALES & SIGNIFICANCE TESTING TYPE OF DATA SCALE Categorical Data Nominal Ordinal Quantifiable Data Interval and ratio PERMISSIBLE TESTS (e.g.) For Categorical Data: Chi square For Quantifiable Data: T-test Correlation coefficient … it is vitally important to think about the way in which the data will be analyzed and presented at the design stage! PAC RESEARCH METHODOLOGY Quantitative Methods 1 SAMPLING CENSUS V. SAMPLE Cost Time Accuracy The destructive nature of measurement THE PROCESS OF SAMPLING Define the population Decide on the sample size Define a frame for that population Select a sampling unit Choose a method of selecting that sample Probability Non-probability Define the sampling plan Select the sample CHOICE OF SAMPLING METHOD PROBABILITY Simple random Stratified random Cluster Multi-stage NON-PROBABILITY Convenience Judgment Purposive Quota … the most important thing about your sampling decisions is that they determine the extent to which you can generalize from your findings. CONCLUSIONS In a survey you typically only get one shot. It is therefore important to allow plenty of time to think through all the potential issues of survey design, www.liuxuelw.com/aozhoulunwen/ data analysis and sampling at the outset. Quite a lot of further reading and hands-on practice may be needed… (That’s normal!) Additional resources General: Diamantopoulos, A. & Schlegelmilch, B.B. (1997) Taking the fear out of data analysis, The Dryden Press, London De Vaux, D. (2002) Surveys in Social Research, 5th edition, Rutledge, London Useful on questionnaire design: Google search: Learning spa