Research Method PDF

RESEARCH METHOD Developing a Question Questions emerge from: ○ Personal observations ○ Broader social problems ○ Previous research ○ Theories that strive to offer explanations for relational events 1st step is always ask...

RESEARCH METHOD Developing a Question Questions emerge from: ○ Personal observations ○ Broader social problems ○ Previous research ○ Theories that strive to offer explanations for relational events 1st step is always ask a question For example —> the increase in divorce help increase research in divorce Research can come from all of these 2 broad types of goals: ○ Describe the nature of events —> researchers want to describe what they observe as they naturally occurs —> observe pattern ○ Establish causal connections —> trying to determine which events have a meaningful impact on other or it may not Different studies do have different goals we should judge/criticize it as such and not as Resist from judging/ jump to conclusions from studies Obtaining Participants 1. Convenience Samples anyone who is ready to participate —> ex. Uni student —> sometimes they are required to participate in research from their course work Always allow 2. Representative Samples —> try to ensure that collectively their participants represent the entire population of their research Issues to keep in mind: - Volunteers may differ from population - WEIRD samples (e.g., Henrich et al., 2010) -western educated industrialized rich democratic (itocracy/) might behave diff than people from other people from diff culture On the other hand it should not matter who it was based on —> it should apply across demographic group research suggest that people all over the world ___ beauty standards convinient samples might not be Correlational Patterns - correlation describe - correlation does not equal causation - natural occurring events —> correlation that are observe are - we don’t know how or why it might change 1. Positive correlation a. change in one event is accompanied by another —> if one goes up so does the other if one goes down so does the other 2. Negative correlation a. inverse correlation —> move in opp direction on dec other inc 3. No (zero) correlation Experimental Design Experiments provide straightforward information about causes and effects because experimenters create and control the conditions they study ○ Manipulate one or more (X or independent) variables and randomly assign participants to the different conditions - how that change affects people making comparison in a dependent variable —> if one thing change what happens to the other ○ Measure and compare groups on (Y or dependent) outcomes/variables - - X caused the ___ in Y - clear more definitive design - has to be able to manipulate —> Data: Self-Reports nature of our data - what type of info/data are we collecting - observing specific 2 types of measure - peoples report - carelful Ask people about their experiences ○ straight to the source and ask people themselves Variety of formats ○ questionare/ verbal interviews/ diaries —> common theme, people are telling us they are not being observe Exact nature may vary Can tell us about meaning ○ Of relational events —> more detailed, get inside peoples head, to understand more invaluable Efficient, and inexpensive ○ expensive access things like equipment or labs —> could be online now Problems with Self-Report Data Interpretation of the questions (Wiederman, 2004) ○ bcs self report occurs what the researchers mean can be misleading ex. How many people have you had sex with? Half consider complexity we might not be aware of —>  Difficulties in recall or awareness (Aicken et al., 2013) ○ faulty memory is a major problem —> debate on how accurately people can remember and report —> people will be the most accurate when they describe specific event that happened recently —> ○ ex. Looks don’t matter to women as much as men but according to research during speed Bias or distortion in reports (e.g., Fisher, 2013; Mitchell, 2010) ○ reluctant to tell researchers anything that will make them look bad present themselves in ○ bias is always a concern —. Try to observe it directly ○ Data: Observations Measure behavior or train people to make observations ○ Accurate, reliable and often very detailed Observe direct ongoing behavior, or use recordings (observation are made later on ○ E.g., Experience-sampling (uses intermittent Can’t access people’s perceptions May be expensive Can suffer from reactivity —> people react