Ch13: Desccriptive Research Strategy Flashcards
What is observational research?
descriptive research that involves observing and recording behaviour
Important issues to consider in observational research:
- Hawthorne effect: demand characteristics/reactivity/observation bias. subjects change behaviour because they know they are being watched.
- Identification of categories
- inter-rater reliability: both observers are observing things consistently
Ways to avoid Hawthorne effect?
- observe public behaviours in public places
- habituate participants to being observed
What are behaviour categories?
- categories of behaviour made by observers to reduce subjectivity
- prior to making observations, researcher identifies categories of behaviour we want to observe (eg., group play, play alone, aggression, social interaction).
- list exactly which behaviour count as examples of each category
- these categories establish a clear operational definition of each construct being examined
Ways of QUANTIFYING observations
-FREQUENCY METHOD: number of instances that a behaviour occurs
-DURATION METHOD: how long does a behaviour last?
INTERVAL METHOD: divide period of observations into intervals and record whether behaviour occurs in each interval (yes or no)
In the frequency and duration methods of quantifying observation, are the same behaviours being measured?
yes, but they are being quantified differently
measuring whether a child smiles during each 1-minute interval following mom’s return is an example of what method of quantifying observations?
interval method
Ways of RECORDING observations?
- TIME SAMPLING: observe for a specified period then record observations. Decide timeline. Watch, record, repeat. Stop at end of period.
- EVENT SAMPLING: for interval 1, looking for behaviour 1. Interval 2, behaviour 2, etc.
- INDIVIDUAL SAMPLING: looking at 1 person at one interval, and then another at the next interval. Group of people, observe 1 at a time. interval 1–>person 1. Interval 2 –> person 2
Ways to OBSERVE behaviours?
- NATURALISTIC OBSERVATION: less interaction, observe from afar.
- PARTICIPANT OBSERVATION: researcher participates in the same activities as the subjects.
- CONTRIVED OBSERVATION: behaviour in a specifically designed setting/situation
Dian Fossey observing gorillas is an example of what kind of observation?
naturalistic observation
The strange situation for observing infant attachment styles is an example of what kind of observation?
contrived observation
a case study is a form of ____ research
observational
In what kind of instances are case studies used?
- typically in rare or unusual cases.
- sometimes used for counterexamples
Your are interested in whether blinking rate is associated with anxiety. You record the number of blinks for a participant during a 30-second period they are either reading a story about bicycle thefts on campus vs reading about a new campus art exhibit.
- what method of observation are you using?
- how are you quantifying your observations?
- what observation sampling technique are you using?
- method of observation: contrived
- quantifying observations: frequency
- sampling technique: time-sampling
What problems in the following example:
A group of researchers is interested in the effort put towards training of dog owners with different sized pets. They go to a dog park and obtain consent from their participants. They decide to observe one dog owner per minute. Researcher 1 records the amount of time
spent correcting behaviour (e.g., scolding) by dog owners as indicating they are putting effort into training. Researcher 2 records the amount of time showing affectionate behaviour (e.g., petting). Researcher 1 concludes that larger dog owners put more effort in,
and researcher 2 concludes the opposite. What are the biggest
issues here?
- Hawthorne effect
- inter-rater reliability
- both observers are observing different beahviours. Should be the same.
Method of observation: naturalistic
Sampling technique: individual sampling
Quantifying technique: duration method
Self-report measures of behaviour, opinions, personality characteristics, etc. falls into what kind of research design?
Survey research design
Types of survey questions:
- demographic questions (eg., what is your age, income, occupation, etc?)
- open-ended questions (allow participants to respond in their own words. eg., what do you think about the current availability of food on campus?)
- restricted questions (presents participants with a limited number of response alternatives. Eg., multiple choice)
What are some issues with rating scales?
- people tend to avoid extremes
- sometimes people select “neutral” for everything
What is a Likert-type scale?
-a rating scale with an extreme on each end, neutral in the middle. Equal spacing between numbers (eg., strongly disagree, disagree, slightly disagree, neither agree nor disagree, slightly agree, agree, strongly agree)
What is a semantic differential rating scale?
semantic differential presents pairs of bipolar adjectives (such as happy—sad, boring—exciting), and asks each participant to identify the location between the two adjectives that best describes a particular individual
A pain scale in the ER is an example of what kind of rating scale?
non-verbal
Pitfalls of survey questions
UNNECESSARY COMPLEXITY: -jargon -vague terms -overloading working memory DOUBLE-BARRELED QUESTIONS -asking two things at once LOADED QUESTIONS -direct responses a certain way NEGATIVE WORDING -agreeing with the question means disagreeing with its content
What survey question pitfall can be identified in the following:
To what extent do you agree that professors should not be required to take daily attendance?
negative wording
What survey question pitfall can be identified in the following:
I enjoy studying and spending time with friends on weekends
double-barreled
What survey question pitfall can be identified in the following:
Do you support the legislation that would unfairly tax hard-working farmers?
Loaded question
What survey question pitfall can be identified in the following:
Do you believe the relationship between cellphone behaviour and consumption of fast food is orthogonal?
Unnecessary complexity
What survey question pitfall can be identified in the following:
Restaaurants should not have to be inspected each month
Negative wording
What are response sets?
-answering all Q’s in a particular manner (eg., yea-saying and nay-saying. Social desirability)
How can you identify whether a person has simply been responding “yes” or “no” only on a survey?
-by including reverse-scored items
One way to prevent responses based on social desirability from occurring:
insert sensitive questions in the middle of the survey
How are surveys conducted?
- questionnaires
- interviews
Researchers are interested in the different methods teachers use to maintain control in junior high classrooms. They go into four 8th grade classrooms, each with approximately 25 students, and record how many times teachers have to discipline students within a one hour period. They notice that the teachers rarely have to correct behaviour from their students and conclude that junior high students are generally well-behaved without any special methods.
- ◦ What kind of observation is this?
- ◦ What method of quantifying & sampling?
- ◦ What could be an issue with this study?
- observation: naturalistic
- quantifying: frequency method
sampling: time sampling - Hawthorne effect
