W1: Research Workshop Flashcards
Types of research methods in health psychology (2)
Quantitative
Qualitative
Quantitative research methods (2)
Measures “how much”, measures numbers
Traditional methods and designs are quantitative, emphasise reliable measures in controlled experiments or surveys.
Qualitative research methods (2)
Understanding “why, what, when, and how” - understand processes at an individual level.
Use interviews, focus groups, narratives, or case studies to explore health and illness concepts and experience -want in-depth person-focused data.
Quantitative research methods - Cross-sectional
What is cross-sectional design?
Measures taken at one time point
Criticisms of cross-sectional design
Has issues with causality and chance of a confounding variable in this design
Quantitative research methods
Longitudinal design what it is?
- Measures taken at two or more time points - e.g. morning or evening or different years.
Quantitative research methods
Longitudinal design overcome the limitations of causality in cross-sectional design as
It measures different time points and removes the possibility of a confounding or third variable.
Quantitative research methods
Longitudinal Disadvantages (3)
Participants may drop out
Expensive to conduct
Time consuming
Quantiative research methods
Experimental
Researchers randomly assign participants to one condition, treatment or group
Quantiative research methods
Quasi-experimental (2)
Group of interest cannot be manipulated, occurs naturally (pre-existing) or unethical to manipulate
Such as genders, sex, race, ethnicity, socioeconomic status, tobacco, alcohol or drug use (commonly used in health to figure out what populations are at risk of poor health)
Limitations of Quasi-Experimental Designs (6)
Lack of control or comparison group as there is no manipulation in the group, potentially unreliable results because of this.
Certain predispositions or individual differences may have influenced categorisation in a group (e.g., socioeconomic status)
- Is behaviour due to belonging to a group or vice versa?
- Is someone an alcoholic because of their low SES or is it the other way around?
- Hard to argue for cause and effect (but we can for experimental)
Recruitment of certain populations may be difficult
Quantitative methods
Advantages of quasi-experimental designs (2)
You might be interested in pre-existing groups - like studying cannabis users.
- e.g., race/ethnicity, sex, gender, socioeconomic status etc..
Quantitative methods
Disadvantages of quasi-experimental designs (4)
Selection differences - any differences between pre-existing groups not controlled or accounted for by the researcher
- Why is someone in that group to begin with - why is someone’s SES low - we don’t know as there are multiple factors that explain their group memberships?
- . The researcher does not know that a person simply being In this pre-existing group is solely responsible for the findings in the study. Therefore:
Cannot demonstrate cause and effect
Quasi-Experimental
Example 1
Researcher wants to examine differences in drinking habits between those who live in the city versus the suburbs..
IV: People living in city vs suburb residence (quasi-experimental) - not manipulated.
DV: Drinking habits - want to know whether it effect drinking habits.
Outcomes: Those who live in the city report drinking more frequently than those who live in the suburbs!
Does living in the city cause you to drink more alcohol? (5)
- No!
- Confounding variables may influence drinking:
- Personality
- Accessibility to bars
- Greater stress in the city vs. suburbs
Quasi-experimental
in terms of generalisability and cause and effect (2)
Quasi-experimental has higher generalisability as pre-existing groups in world than experimental
but can not demonstrate cause and effect
