Week 10 : Survey Research

Question 1

What is survey research

Answer 1

• uses self-report from respondents
• tries to obtain generalizable samples (ideally large and random)

Question 2

Types of survey research

Answer 2

1. interviews
2. phone surveys
3. questionnaires

Question 3

Types of survey research

Interviews

Answer 3

• structured: have a fixed script
• unstructured: more conversational
• drawbacks… costly, interviwer bias, social desireability concerns

Question 4

Types of survey research

Phone surveys

Answer 4

• structured or unstructured
• cheaper than face-to-face interviews
• fewer social desirability concerns
• used to be easy to get random samples but cell phones and telemarketing kinda ruined that

Question 5

Types of survey research

Questionnaires

Answer 5

• paper or electric
• cheapest
• fewest social desirability concerns

Question 6

Questionnaire advantages

Answer 6

• can asses non-observable variables, as well as variables that you cannot (ethically) manipulate… [e.g. demographic info, attitudes & belief, past behaviour, motication, personality traits, etc.]
• easy to administer & score
• can gather a lot of info
• requires few resources

Question 7

Questionnaire disadvantages

Answer 7

• accuracy may be low cuz… participants may lack insight abt certain variables, may forget previous behaviour & may respond in a socially desireable manner
• since we are not manipulating the IV, we cannot demonstrate causation… this is true of all correlational/non-experimental research

Question 8

Constructing good surveys

a good survey item is…

Answer 8

• BRUSO
  1. brief
  2. relevant
  3. unambiguous
  4. specific
  5. objective

Question 9

BRUSO

Brief

Answer 9

• avoid long or run-on sentences
• avoid unnecessary words
• avoid technical terms, acronyms, jargon

Question 10

BRUSO

relevant

Answer 10

• avoid the temptation to include lots of extra iems ‘just in case’
• especially personal ‘nosy’ questions

Question 11

BRUSO

Unambiguous

Answer 11

• avoid vague or imprecise terms…. e.g. ‘health problems’, ‘psychological issues’
• avoid negative wording… ‘do you disagree with the idea that parents should not spank their children?’

Question 12

BRUSO

Specific

Answer 12

• avoid double-barreled questions… questions that ask 2 things at once
• e.g. ‘are the big macs at dons delicious and healthy’`

Question 13

BRUSO

Objective

Answer 13

• avoid leading questions, emotionally charged words… e.g. are you in favour of eliminating the wasteful expenses in the budget?
• questions should be neutral… e.g. are you in favour of reducing the budget?

Question 14

Reverse scored items

Answer 14

• it can be useful to use some reversed scored items
• reversing some wording helps prevent yea-saying and nay-saying

Question 15

Closed-ended survey items…

Answer 15

• give a limited number of responses
• must be carefully designed
• risk of missing information
• very easy to score
• closed-ended categorial items… for a categorial question, provide a list of options & have an other
• closed-ended continuous items… use rating scales (usually 5/7 Likert) & decide the anchors

Question 16

Open-ended survey items…

Answer 16

• allow participants to respond however they want
• provides a lot more information…but… effortful for participants
• need to decide how to score responses & this takes time

Question 17

Assembling the survey…

Answer 17

• order matters (item order bias)
• place less sensitive/objectionable items before more sensitive items
• demographic items are rarely biased and thus go last (age, ethnicity, etc.)
• organize in a coherent, visually pleasing format [large font, not too many items per page (reduce scrolling) & viewable on different devices]

Question 18

Simple random sampling

Answer 18

• everyone in the population has an equal chance of participating
• the sample should resemble the population
• almost impossible to obtain (limited by location, recruiting methods, resources)

Question 19

Stratified random sampling

Answer 19

• important subgroups are identified
• obtain a random sample of each subgroup to mirror the population

Question 20

non-random sampling

Answer 20

• everyone in the population does not have an equal chance of participating
• may lead to a biased sample, often due to selection bias

Question 21

non-random sampling

Convenience sampling

Answer 21

• use participants who are easily available
• very common
• limits external validity
• student populations… 81% of research participants are university undergraduates
• internet populations… low paying online research
• voluntary participation… volunteer bias (more women)