data type

data type

1. secondary

   • internal
     • eg. p&l statements, inventory records, sales figures, customer database, online behaviour tracking data
   • external
     • eg. govt reports, market research database, academic research database, industry leader reports
   • adv
     • efficiency, cost, validity and objectivity
   • disadv
     • recency , comparability

2. primary

   • qualitative (exploratory)
     • interviews
       • in-depth
         • characteristic
           • semi-structured qns, face-to-face
         • technique
           • interview guide, comfy env, sufficient time to think and answer, summarise their ans, avoid evaluate comments and leading qns, don't interrupt
             • general to specific, factual to attitudinal
       • projective
         • metaphors, analogies, third party projections (eg. what do your friends think), role playing
       • analysis
         • themes, frequency count, polarity
     • focus groups
       • homogeneous subjs, moderator
     • observations
       • reasons
         • consumers can not accurately recall their behaviour
         • consumers are unwilling to admit their behaviour
       • conditions
         • observable, short time interval during occurrence, public setting
       • observable stuff
         • 1. physical action
         • 2. expressive behaviour
         • 3. verbal behaviour
         • 4. temporal behaviour
         • 5. spatial relationships and locations
         • 6. physical objects
       • categories
         • direct/ indirect
           • behaviour today vs past
         • disguised/ undisguised
           • mystery shopper vs research
         • human/ mechanic
           • eye tracking monitor, voice pitch analyser
         • structured/ unstructured
           • pre-selected vs random events
   • quantitative
     • descriptive (surveys)
       • types
         • cross-sectional
         • longitudinal
       • adv
         • efficiency, standardisation, easy to interpret and analyse data, subgroup differences
       • surveys
         • person-administered
           • face-to-face
           • telephone
           • adv
             • instant feedback, quality control, flexibility
           • disadv
             • questionnaire needs to be short, high cost, social desirability, subjects not representative
         • mail/ online
           • adv
             • relatively low cost, little social desirability bias, easy follow up, more questions can be included
           • disadv
             • low quality control, slow data collection process, low response rate, high non-response question rate, possible misunderstanding of qns
         • selection method factors
           • situational
             • budget of available resources, completion time frame, quality requirement of the data
           • task
             • length of the questionnaire, stimuli to be presented in the survey
           • subject related
             • diversity of subjects, incidence rate, degree of survey participation
         • errors
           • sampling error
             • wrong sample selected, small sample size
           • non-sampling error
             • respondent (hardest to control)
               • response rate
                 • no. of completed surveys / no. of eligible units in the sample
               • no. of eligible units in sample
                 • completions + (refusals + not reached) x eligible rate
               • eligible rate
                 • completions / completions + ineligible
               • unconscious misrepresentation
                 • when subjects unconsciously give false answers
                 • extremity bias
                   • subj use extemes when responding qn, i HATE this product
                 • misunderstanding, guessing, attention loss, distraction, fatique
                 • mental set error
                   • answers to the previous qn influence answers to a later qn
             • research design
             • research administrator
     • causal (experiments)
       • definitions
         • independent variable
         • dependent variable
         • extraneous variables (noise)
       • types
         • laboratory experiment (artificial setting)
         • field experiment (test markets)
       • conditions for causality
         • covariation
           • the DV changes as IV changes
         • time sequence
           • IV must occur either prior to or simultaneously with the DV
         • systematic elimination
           • all other possible causal variables (extraneous variables) are eliminated from the r/s
       • experimental design
         • after-only design
           • X O1 - single group of test units is exposed to a treatment x, single measurement on the DV is taken, do not properly control for the effects of extraneous variables
         • one group, before-after design
           • O1 X O2 - a group of test units is measured twice, treatment effect E = O2 - O1, validity of conclusion is questionable, extraneous variables largely uncontrolled
         • after-only with control group
           • EG (R) : X O1 | CG (R) O2 - the experimental group (EG) is exposed to the treatment, and the control group (CG) is not, measurements on both groups are made only AFTER the treatment
             • randomisation R - subjects assigned to different treatment and control conditions randomly to average out the effect of extraneous variables
         • before-after with control group
           • EG (R): O1 X O2 | CG (R): O3 O4 - the EG is exposed to the treatment, and the CG is not, a pre treatment measure is taken on each group, the treatment effect E = (O2 - O1) - (O4 - O3)
         • validity in experiments
           • internal
             • extent to which the change in the DV was actually due to the IV. control of extraneous variables is necessary condition
               • lab> field
           • external
             • extent to which cause-and-effect r/s found in exp can be generalised in to the population
               • lab < field
             • threats
               • sample does not represent the population
               • artificial setting in lab exp
             • solutions
               • employ correct sampling plan, increase sample size, replicate in field exp
           • cost and implementation difficulty
             • lab < field