• Cause and effect – With an experiment, we can determine cause and effect because we are systematically manipulating variables – Hypothesis-what do we think will happen in the study?
– Independent variable(s): – Dependent variable(s):
• Operational definitions – What do you mean by the I.V.?
– What do you mean by the D.V.?
– You need to define them so people who read your study know what you are talking about
• Subjects: – Population – Sample - #s – Random Assignment – Experimental Group – Control Group
• Extraneous vs. Confounding variables – Extraneous – – Confounding –
• 1. Internal – The problem with internal validity is that it becomes very difficult to generalize the results to the rest of the population. (This can also be a problem with case studies).
• 2. External validity-
• 3. Sampling bias • You need to draw a large enough sample to represent the population under study and then utilize random sampling
• 4. Placebo effects – • Simply placing something in one’s mouth may cause one to experience a placebo effect • Placebo effects can ruin the experiment because you cannot tell if the drug actually had an effect on the subject or not.
• 5. Distortion of self-report-Occurs when interview and surveys are given. The subject may lie in giving their answers (or just stretch the truth). Can happen in one of two ways: • A. Social desirability • B. Response set (Ex- when people agree with or disagree with everything on the survey. Most surveys are set up to elicit positive/negative answers or so that different choices should be made.
• 6. Experimenter bias • The best way to overcome this is to do a double-blind study, in which the experimenter does not know which subjects are in which groups
• In order to study development of human beings, we use several different designs: – 1. Cross-sectional – 2. Longitudinal – 3. Sequential
• 1. Naturalistic Observation – Pros – – Cons –
• 2. Laboratory Experiment – Pros – – Cons –
• 3. Surveys – Pros – – Cons –
• 4. Interviews – Pros – – Cons –
• 5. Case Study – Pros – – Cons –
• 1. Subject participation – voluntary – Should not be forced or feel pressured – Should be informed of any factor of research that might make them decide not to participate – Should be allowed to end participation at any time and still get paid (if they are getting paid)
• 2. Subjects are not to be harmed in any way – Protect from psychological and physical harm – Studies that may only effect one’s emotions in a slight way are acceptable
• 3. Deception of participants – Is it okay?
– Should be done to inform participants – Debriefing of subject is usually necessary
• 4. Right to privacy should not be violated – Don’t give out information on your subjects – If data regarding subjects will be released, subjects must be informed and consent must be obtained
• 5. Can we cause them harm?
– Yes to animals – but you must be able to justify it – Animals should be maintained in a manner that is considered decent treatment
• 6. Approval must be obtained from the institution that the researcher works for – Research must be reported in an ethical fashion – Research should be reported as soon as possible for verification
• Mean-Average of all the data • Mode-# that appears most often. You may have a bimodal distribution which means that two #’s appear the same # of times and are the most frequent #’s in the data set.
• Median-# in the middle • Range – subtract the lowest # from the highest #
• While experiments tell us about cause and effect, • Usually after the fact – correlational studies do NOT tell about causation • Often cannot experiment on humans – they ONLY tell us if there is a relationship • Follow-up may be experiment with animals
• Look for: – Positive vs. negative relationships – Strength of the number
• +1 –
• +.7-.9 – • +.4-.6 – • +.1-.3 – • 0 –