Class Notes (837,553)
Canada (510,314)
Psychology (5,220)
PSYCH 1X03 (1,058)
Joe Kim (989)
Lecture

Lecture 2+3 Research Methods.docx

9 Pages
38 Views
Unlock Document

Department
Psychology
Course
PSYCH 1X03
Professor
Joe Kim
Semester
Fall

Description
Psychology Online Lecture 2/3: Research Methods Bernard Ho September 16, 2010 Basic Rules of Scientific Inquiry  Parsimony o When presented with two otherwise equally valid explanations, scientists tend to prefer the simper of the two o However, if one explanation can better account for the known facts, then the more comprehensive explanation is likely to be adopted o The principle of parsimony is specifically applied to the situation of competing explanations that do an equally good job of accounting for the known facts  Natural order o As far as possible, we will attribute the same effects to the same causes o We assume that a behaviour (smiling) is a reflection of the same underlying mechanism in all people all over the world o It is the same root mechanism (happiness) that produces smiles on lottery winners and proud grandparents everywhere o This principle is only to be applied in situations comparing the same effects  Generalizability o The same causes that produce our effects in the lab also produce those effects in everyday life situations over which the scientist has no control  Conservatism o Scientists are conservative in the sense that they tend to support the current theory until new facts accumulate to force the theory to be modified or abandoned  Empiricism and Objectivity o Modern scientific methodology is based on empirical data following the view that knowledge should be based on actual observation and not on reason alone o Not to say that logic and reason have no place in modern science, they play key roles o We certainly cannot carefully observe everything, we must pick and choose where to direct limited resources  Experience and common sense vs. scientific inquiry o One of the main dangers of common sense is the phenomenon known as hindsight bias o When we learn something new, we have a tendency to think that we could have guessed it, or we already partially knew it o It is this bias that can allow any reasonable statement to be categorized as common sense o Any explanation makes sense and seems like common sense to us, even when the information is new or not intuitive Scientific Method 1. Construct a theory a. General set of ideas about the way the world works b. Begin by studying the existing collection of information about the world, such as previous work published by other scientists c. Operation definition i. Describes the actions or operations that will be made to measure or control a variable in any given study or experiment ii. Without such an operation definition, it would be possible to consider a wide range of behaviours to be indicative of any terms under different circumstances 2. Generate a hypothesis a. Testable statements guided by theories that make specific predictions about the relationship between variables 3. Choose research method a. The way in which a hypothesis will be tested b. Methods such as experiments allow scientists to collect data about how the events of the world unfold, which may or may not be in line with their hypothesis c. Scientists analyze the data to note any specific trends or relationships that the research has revealed, ultimately leading to the decision to accept or reject the original hypothesis 4. Collect data a. Taking measurements of the outcome of the test using one or more of several techniques 5. Analyze data a. Understand the data and discover trends and relationships between variables b. Ultimately leading to the decision to accept or reject the hypothesis 6. Report findings a. Publish articles in scholarly journals b. Individual scientists and scientific community as a whole review all findings on a topic to revise existing theories that define our current understanding of the world c. Undergoes a rigorous referee process to ensure research is scholarly, accurate and meaningful to the field 7. Revise existing theories a. To include new information into our understanding of the world b. Paradigm shift i. Particularly dramatic change in our way of thinking  Provides a standardized process for scientific research  Minimizes biases, conflicts and other problems to promote accurate results and scientific discourse Conducting an Experiment (hypothesis)  Eric sees an infomercial for “MegaStudy”, which enhances memory and helps focus during studying  Theory  Test performance can be affected by external factors  Hypothesis  Students taking energy drink should show higher test scores compared to those not taking the drink  Anecdotal evidence  Evidence gathered from others or self experience  If he gets good grade  drink worked  If he fails  drink fails  Problems with testing a hypothesis  Single experience may not be representative of the general result that would occur if it was repeated several times  Can't be sure your experience is the same as what others would experience under the same circumstances  Can’t be sure that any change in test performance was due to the drink at all (easier/harder test, studied more etc.) Control Groups  Experiment (choosing research method)  Scientific tool used to measure the effect of one variable on another  A variable is anything that is free to take on different values  Scientist manipulates independent variable and observes the effect the manipulation has on the dependent variable  Eric has not properly manipulated the independent variable  Control groups  Eric will take a test after drinking the energy drink, but he will ask his roommate to take the same test without taking the drink  However, what if Eric is smarter than his roommate?  In simplest form, contains two groups  Experimental group o Receive manipulation of independent variable during the experiment  Control group o Does not receive manipulation  Participants in both group should be as similar as possible o Eric might try to find a student who gets the same mark as him, but also who studies similar to him and takes the same classes  Within-subject design  Tests the same subject repeatedly while the independent variable is manipulated  Experimenter is comparing the performance of the same group of subjects within the same study under different conditions  Eric takes several tests during the semester, some he drinks before, some he doesn’t  Can compare the performances on those tests after  He is acting as his own control group  Can be time consuming and costs money  Test might increase in difficulty  This variation has the benefit of minimizing the potential confounding variable, but may leave open the possibility of a practice effect  A participant’s performance could increase because he/she becomes more experienced, called practice effect  Between subjects experiment  One group acts as experimental group and another acts as a control group  Experimenter is comparing the performance on the dependent variable between the subjects of the experimental and control groups  This design can lead to relatively straightforward cause and effect conclusions  Important for individuals being tested to be as similar as possible in every way except in the manipulation of the independent variable  A systematic difference between subjects other than the independent variable is called the confounding variable or extraneous variable Sampling  Selecting subjects (collecting data)  Although Eric decides to pick subjects that are all similar (Dutch girls with an average of 93%), he is limiting the scope of his conclusions  The results may indicate that the drink improves the performance of these type of people, but it may not be so for other races or genders  Population  General group of people we are trying to learn about  Sample  Members of the population that we collect data from  Best technique to pick subjects is by random sample  Then random assortment, by randomly assigning subjects to either experimental or control groups, to avoid biases  Every member of the population must have an equal chance of being selected for the sample  If we design an experiment with two or more groups of subjects that will receive different treatments to be compared to one another, it is equally important that we use random assignment of subjects to groups  In other words, which subject ends up in which group must be randomly determined  As with random selection, random assignment can help us avoid systematic errors with extraneous variables  When we draw inferences about a population, it is important that the inferences we draw be based on a sample that is representative of the population  We deem a sample as being representative if its demographic makeup is similar to that of the population we are interested in studying Conducting an Experiment  When data is collected and analyzed, two specific errors can be made o Type 1 error  Researcher has used an inappropriate inferential statistical test to erroneously reject the null hypothesis and conclude that there is a statistically significant effect  Even if a result is concluded to be statistically significant, there is still a 5% chance that the difference was found due to chance o Type 2 error  The researcher erroneously fails to reject the null hypothesis and misses out on an existing underlying effect  Can occur for several reasons 
More Less

Related notes for PSYCH 1X03

Log In


OR

Join OneClass

Access over 10 million pages of study
documents for 1.3 million courses.

Sign up

Join to view


OR

By registering, I agree to the Terms and Privacy Policies
Already have an account?
Just a few more details

So we can recommend you notes for your school.

Reset Password

Please enter below the email address you registered with and we will send you a link to reset your password.

Add your courses

Get notes from the top students in your class.


Submit