Presentation is loading. Please wait.

Presentation is loading. Please wait.

Module #3: Research Strategies

Similar presentations


Presentation on theme: "Module #3: Research Strategies"— Presentation transcript:

1 Module #3: Research Strategies

2 Research Strategies We will look at 3 different types of research methods: 1. Descriptive a. case studies b. surveys c. naturalistic observations 2. Correlational 3. Experimental

3 Descriptive – Case Study
Case Study – studying one individual in detail, thinking it will lead to information about all people. Jean Piaget (Cognitive Stages of Development) studied children to discover information about learning. Case studies sometimes were misleading because there are always exceptions. The one being studied might be unusual and does not tell us about human behavior.

4 Descriptive – Case Study
Case Study – Genie Wiley: Feral Child Feral – lived away from human contact Genie was the victim of severe abuse. She was locked alone in a room for the first 13 years of her life strapped to a toilet or bound to a crib with her arms and legs tied. She was never spoken to and thus never acquired a language. Upon rescue she gradually started to acquire language skills.

5 Descriptive – Case Study
She lived at the institution where she was being studied for 4 years. After leaving institution she bounced around different places where she experienced more abuse and her language skills began to regress. In 2008 it was reported that Genie was living in California in psychological confinement, speechless.

6 Descriptive – Surveys Surveys – Ask people to report on their behavior/opinions. Less depth than case studies but are asking larger number of people. Wording can have a major effect in surveys and can have a big affect in answers given. “Do you believe in helping the needy?” Vs. “Do you support welfare programs?”

7 Descriptive – Surveys False consensus effect – We surround ourselves with people that have similar habits and attitudes. This leads us to assume everyone thinks and acts like we do. Representative sample is to randomly survey people from all backgrounds and experiences. Population – everyone you want to study and describe. Random sample – a sample that fairly represents the population.

8 Descriptive – Surveys Example: Conducting a survey on the theme for Senior Prom. Population: Senior Class Random Sample: The seniors in this class DO NOT represent a random sample. You would need representatives from all the various levels of classes and groups. (Reg, Hon, athletes, musicians, etc.) Large samples are better than small ones, but must be representative.

9 Descriptive – Surveys Accuracy of data is an issue because people sometimes distort their answers or fail to recall information correctly. In the 1970s, researchers noticed a lot of babies were born with deformed limbs. They found the strongest relationship between the mothers was the drug thalidomide during pregnancies. Controlled experiments with rats verified the drug caused abnormal limb development in babies.

10 Descriptive – Observations
Naturalistic Observations – watching and recording behavior in their natural environment. Does not explain behavior, it describes it. Jane Goodall – #1 expert on chimpanzees. Conducted a 45 year study of wild chimps in Tanzania, Africa.

11 Correlation Correlation – When two things have a relationship and are dependent of each other. It allows one to predict the other. Correlation Coefficient – How much two things are correlated. From -1 to +1. Strength of correlation. Higher number means a stronger correlation. Direction of relationship: positive or negative. r = +0.45

12 Correlation Positive Correlation – Two things rise and fall together.
In this example, the more years of education you have, the older you are when you have your first child.

13 Correlation Negative Correlation – Inverse relationship. When one rises, the other falls. Weak correlations have coefficients near 0. In this example, the more hours of video games played, the lower the GPA.

14 Correlation Correlation DOES NOT mean causation.
Because two things are correlated, does not mean they are caused by each other. Among men, length of marriage is positively correlated with hair loss. This does not mean that marriage causes hair loss. Correlation shows a POSSIBLE cause-effect relationship, but does not prove it.

15 Correlation Illusory Correlation – When we think two things are correlated, we look for things that confirm our belief even if they aren’t really correlated. “Being wet and cold does not cause a person to catch a cold.” It only seems to have a correlation. We notice random coincidences and assume there is a correlation, even when there is no correlation.

16 Experimentation Experiments allow researchers to control conditions to isolate cause and effect. Experiments manipulate one or more variables, while holding constant other factors, to observe the effect it has on another variable. An experiment manipulates a factor to determine its effect.

17 Experimentation Placebo – a drug that has no medical value given to deceive a participant into thinking they are receiving an actual treatment. Used for control groups. Double-blind procedure – when both the participant and the research staff are unaware of who received a placebo. Placebo Effect – improvement of medical condition when given a placebo. Believing you are getting a treatment can reduce symptoms.

18 Additional Terms Operational Definition – describes the specific procedure used to determine the presence of a variable. Ex. If conducting an experiment and are looking for a reaction, you would operationally define what reaction is. It could be wide eyes, open mouth, verbal exclamations, shocked silence, frightened movements, etc.

19 Additional Terms Confounding Variables – differences between the experimental and control groups NOT caused by the independent variable. Ex. You conduct an experiment to find out if drinking orange juice helps students stay awake. You give half of the students juice and the other half did not receive juice. The ones who drank the juice stayed awake. What are the confounding variables?

20 Additional Terms Experimenter Bias – A phenomenon that occurs when a researcher’s expectation or preferences about the outcome of a study influence the results obtained. Ex. If you think that the drug in your study works really well, you might ask participants who took the drug, “You feel better, don’t you?” If you think watching TV negatively effects grades, you put all of the lesser students into the TV group.

21 Additional Terms Response Bias – A phenomenon that occurs when a participant answers questions in the way they think the questioner wants them to answer instead of their true beliefs. Ex. (True Story) Mr. Yolich asks his son which was worse, striking out in a game or being yelled at by him. His son responded, “being yelled at”. Mr. Yolich asked again. This time his son answered striking out because that was what Mr. Yolich wanted him to say.

22 Experimentation Experimental Condition – A group of people who receive an actual treatment. Control Condition – A group of people who do not receive the treatment. Allows comparison of two groups as to the effect of the treatment. Participants must be assigned randomly (distributed equally amongst experimental and control groups by chance).

23 Experimentation Independent Variable (IV) – The experimental factor that is being manipulated. The variable whose effect is being studied. Dependent Variable (DV) – the variable that may change depending on the independent variable. Outcome of changing the IV. Experiments manipulate the IV, measure the DV, and control all other variables. Experiments test the effects of IV.

24 Experimentation “Experiment on how much a plant grows based on how much water it receives.” Independent Variable – Amount of water plants received. Dependent Variable – The growth of the plants. The growth of the plants depends on how much water the plants received.

25 Statistical Reasoning
Measures of central tendency – mean, median, and mode. Mean – Average (sum of scores divided by number of scores) Median – Midpoint (Order scores and find the middle) Mode – Most frequent (Score that appears the most) Range – The difference between the lowest and highest scores.

26 Statistical Reasoning
Skew – When unusually high or low scores distort the mean. No Skew Skewed!

27 Statistical Reasoning
Standard Deviation – How much scores deviate from the mean. 1. Find the mean of the data. 2. Find out how much each data point deviates from the mean. 3. Square the deviations. 4. Find the mean of the squared deviations. 5. Find the square root of the mean.

28 Statistical Reasoning
Score Deviation Squared from Mean Deviation 70 71 85 90 -9 -8 +6 +11 81 64 36 121 3 2 316 1 Sum of dev = 302 316/4 = 79 (mean) 4

29 Statistical Reasoning
Sum of dev. # of scores Standard Deviation = 4 5 302 4 = Standard Deviation = 8.69

30 Standard Deviation The lower the standard deviation, the more closely packed the scores are. This means they are not as scattered apart. Low SD. Data is tightly packed. Both have same mean. High SD. Data is spread far apart.

31 Statistical Reasoning
Statistical Significance – a statistical statement of how likely it is that a result occurred by chance. If sample averages are reliable and the difference between them is relatively large, then it has statistical significance.


Download ppt "Module #3: Research Strategies"

Similar presentations


Ads by Google