Help
Options
Didn't know it?
click below
click below
Knew it?
click below
click below
Don't Know
Remaining cards (0)
Know
retry
shuffle
restart
0:00
Embed Code - If you would like this activity on your web page, copy the script below and paste it into your web page.
Normal Size Small Size show me how
Normal Size Small Size show me how
Research Methods
| Question | Answer |
|---|---|
| Two types of Research | 1. Experimental 2. Non-Experimental |
| Experimental Research | Only this can prove cause and effect. Performed in a lab. Control variables. |
| 7 steps to the Scientific Method | 1. Hypothesis 2. Literature Review 3. Design 4. Carry out experiment 5. Stats analysis 6. Results and Conclusions 7. Share |
| Hypothesis | Usually designed from a theory. What question do you want answered? Turn it into a statement. |
| Literature Review | Research to examine other research about your hypothesis. Is it valid? Can you replicate other studies? |
| Design | Randomly recruit population Randomly assign to Experimental or Control IV given to only Experimental Group No IV for Control Group Measure DV for both groups Descriptive Stats: mean/median/mode/st. dev Inferential Stats; P=<.05 Cause & Effect? |
| Random Selection | Anyone has the same chance to be in your study. Choose randomly from those who wish to be a participant. |
| Random Assignment | Randomly assign your population to the Experimental or Control Group. How? Choose every other name, pick out of a hat, etc. |
| Conduct Experiment | Experimenter--someone who can run the subjects through the study. It should not be you! |
| Demand Characteristic | Experimenter gives off a clue by accident, and the participant thinks they know the study's purpose and gives correct answers |
| Placebo Effect and Halo Effect | Placebo: experimenter gives off clue and participant in the control group thinks they know the study's purpose and gives correct answers. Halo: same as placebo only in the experimental group Both effects will skew the results |
| Sampling Bias | When you don't use a large enough population from all walks of life, your results will be biased toward the Pop you used. |
| Quasi Experiment | When you can't randomly assign your population to your experimental and control groups. Example: Doing your research in a school. |
| Single Blind Study | The experimenter knows the study's purpose |
| Double Blind Study | The experimenter does not know the study's purpose |
| Hawthorne Effect | When people being observed know you are watching them and change their behavior. |
| Data Analysis | 1. Descriptive Statistics 2. Inferential Statistics |
| Descriptive Statistics | Mean, Median and Mode and Std. Dev. |
| Mean | Average of your scores |
| Median | Put scores in order. Middle score is your median. If it is a set of even numbers, find the average of the middle 2 numbers. |
| Mode | Which score has the most responses? It can be bimodal or trimodal also |
| Measures of Central Tendency | Another name for the mean, median and mode |
| Standard Deviation | How much do the scores vary from the mean? If you are given the variance, take the square root. That is your st. dev. |
| Variance | The average of how the scores vary from the mean. If you know the standard deviation, you can square that to find the variance. |
| Negative Skew | The curve will have its tail going to the left. Lower level scores are missing so it is skewed (empty) toward the negative side. |
| Positive Skew | The curve will have tail going to the right. Upper level scores are missing so it is skewed (empty) toward the positive side |
| Inferential Statistics | You perform an ANOVA or a T test on your two sets of means, one from the experimental group and the one from your control group. |
| T-Test | An inferential statistic that will compare the difference between two sets of means |
| ANOVA | An inferential statistic that will compare the difference in two or more sets of means |
| Significance | Are your two sets of means significantly different? The p (probability) factor must be equal to or < .05 |
| How do you interpret the p factor? | 5% of the time your hypothesis will not work, but 95% of the time it will. |
| Ethics in Psychology Studies | The harm to the participant is measured against the value of the research. If the harm to the participant because of deception is higher, then the experiment should not take place. |
| IRB | Internal Review Board in a university or research facility that reviews your application to perform the study. They will decide whether or not you can run it. |
| Some unethical experiments which have been performed | Milgram Study Stanford Prison Experiment The Scopes monkey study |
| 5 major guidelines for your experiment's application | Show how you will get the subject's consent What are the conditions of consent? What is the cover story? How will you debrief the participants? How will the Board decide whether your study is moving forward? |
| Cover story of the experiment | What have you told the participants that the experiment is about? Can't tell them the purpose, so you create a cover story This is the deception to the subjects. |
| Debriefing | Every participant must be told the real purpose of the experiment when the experiment is over. Then payment. |
| APA | American Psychological Association drew up the application guidelines for your experiment which are detailed in your application. |
| Decision Making | The IRB will decide the merits of your study--does the harm to the participants because of the deception, outweigh the value of the research? |
| Non-Experimental Research | Not performed in a lab: surveys, observations, interviews, case studies. These methods can only show relationships between variables |
| Data analysis for surveys | 1. Frequency Bar Graphs 2. Scatterplots |
| Frequency Bar Graphs | How many people who completed your Likert scale gave that statement a 1?2?3? and so forth. You can see the most frequent response and least frequent at a glance |
| Scatterplots | Used to compare the scores of two separate statements. We are looking for evidence of a correlation that will illustrate a relationship between two or more variables |
| "r" factor | Shows us if we have a correlation between our variables. The r stands for relationship. |
| How do we know if a correlation is a strong one? | The closer the r factor is to 1.0, the stronger is the correlation. REMEMBER THE SIGN DOES NOT MATTER!! |
| Positive Correlation | The line on your scatterplot will be to the right. This means as one variable increases in value, so does the other variable. |
| Negative Correlation | The line on your scatterplot will be to the left. This means as one variable increases in value, the other variable decreases. |
| Observations | You will perform an observation in a naturalistic setting, and write your results. |
| Interviews | Can be one one one, or with a focus group. An interview should be taped and transcribed for transparency reasons |
| Case Study | Can be a study of one person, or a small entity like a family, regarding an issue you are studying. You can use surveys and observations, but mostly you will be interviewing. |
| Confounding Variables | Variables, like a weight loss program at the same time as your study, that you didn't control for. |
| We operationalize | So others can replicate our study |
Created by:
lmckay
Popular Psychology sets