The Illusory Correlation
phenomenon of seeing the relationship one expects in a set of data even when no such relationship exists. Ex: people sometimes assume that because two events occurred together at one point in the past, that one event must be the cause of the other.
measure of the relation between two or more variables. The measurement scales used should be at least interval scales, but other correlation coefficients are available to handle other types of data. Correlation coefficients can range from -1.00 to +1.00. The value of -1.00 represents a perfect negative correlation while a value of +1.00 represents a perfect positive correlation. A value of 0.00 represents a lack of correlation.
the philosophy of science emphasizes evidence, especially as discovered in experiments. It is a fundamental part of the scientific method that all hypotheses and theories must be tested against observations of the natural world rather than resting solely on a priori reasoning, intuition, or revelation.
A proposition or theory is falsifiable if it is
hypothetically possible for a test or observation to prove it false, regardless of the accuracy of the proposition. A proposal or theory that can never be shown to be false is not falsifiable.
theory or speculation which has the trappings and rhetoric of science, and is presented as science, but is not valid science. Theories are typically classed as pseudoscientific if they fail the scientific method. Pseudoscientific theories are typically not falsifiable, lack objectivity, and their purveyors show unwillingness to allow neutral outsiders to observe, test, or replicate their findings.
process of self-regulation by a profession or a process of evaluation involving qualified individuals within the relevant field. Peer review methods are employed to maintain standards, improve performance and provide credibility. In academia peer review is often used to determine an academic paper's suitability for publication.
To conclude causation, three things must occur. -
1.There is a temporal order of events in which the cause preceded the effect. This is called temporal precedence. Thus, we need to know that television viewing occurred first and aggression then followed. 2. When the cause is present, the effect occurs; when the cause is not present, the effect does not occur. This is called covariation of the cause and effect. We need to know that children who watch television violence behave aggressively and that children who do not watch television violence do not behave aggressively. 3. Nothing other than a causal variable could be responsible for the observed effect. This is called elimination of alternative explanations. There should be no other plausible alternative explanations is very important; suppose that the children who watch a lot of television violence are left alone more than are children who don't view television violence. In this case, the increased aggression could have an alternative explanation: lack of parental supervision.
The politically correct and ethical way to address people in your experiment is
participants Don't use the word subjects.
What is a two tailed t test? -
If you are using a significance level of 0.05, a two-tailed test allots half of your alpha to testing the statistical significance in one direction and half of your alpha to testing statistical significance in the other direction. This means that .025 is in each tail of the distribution of your test statistic. When using a two-tailed test, regardless of the direction of the relationship you hypothesize, you are testing for the possibility of the relationship in both directions. For example, we may wish to compare the mean of a sample to a given value x using a t-test. Our null hypothesis is that the mean is equal to x. A two-tailed test will test both if the mean is significantly greater than x and if the mean significantly less than x. The mean is considered significantly different from x if the test statistic is in the top 2.5% or bottom 2.5% of its probability distribution, resulting in a p-value less than 0.05.
What is a onetailed test-
If you are using a significance level of .05, a one-tailed test allots all of your alpha to testing the statistical significance in the one direction of interest. This means that .05 is in one tail of the distribution of your test statistic. When using a one-tailed test, you are testing for the possibility of the relationship in one direction and completely disregarding the possibility of a relationship in the other direction. Let's return to our example comparing the mean of a sample to a given value x using a t-test. Our null hypothesis is that the mean is equal to x. A one-tailed test will test either if the mean is significantly greater than x or if the mean is significantly less than x, but not both. Then, depending on the chosen tail, the mean is significantly greater than or less than x if the test statistic is in the top 5% of its probability distribution or bottom 5% of its probability distribution, resulting in a p-value less than 0.05. The one-tailed test provides more power to detect an effect in one direction by not testing the effect in the other direction. A discussion of when this is an appropriate option follows.
In order to test whether your hypothesis is true or not, you have to carry out some research to see if you can back it up. So you set up a hi-tech alien detection system and record whether times of alien activity are correlated with when your socks go missing. However, when you get your results, it's possible that any relationship that appears in your data was produced by random chance. In order to back up your hypothesis you need to compare the results against the opposite situation: that the loss of socks is not due to alien burglary. This is your null hypothesis - the assertion that the things you were testing (i.e. rates of alien activity and sock loss) are not related and your results are the product of random chance events.
the loss of my socks is nothing to do with alien burglary.
the loss of my socks is due to alien burglary. In statistics, the only way of supporting your hypothesis is to refute the null hypothesis. Rather than trying to prove your idea (the alternate hypothesis) right you must show that the null hypothesis is likely to be wrong - you have to 'refute' or 'nullify' the null hypothesis. Unfortunately you have to assume that your alternate hypothesis is wrong until you find evidence to the contrary. So it's innocent until proven guilty for the aliens
form a coherent and logically consistent structure that serves two important functions; first theories organize and explain a variety of specific facts or descriptions of behavior. Second, theories generate new knowledge by focusing our thinking so that we notice new aspects of behavior.
summary of the research report and typically runs no more than 120 words in length. It includes information about the hypothesis, procedure and broad pattern of results.
outlines the problem that has been investigated. Past research, specific expectations of the researcher as hypotheses.
how exactly the study was conducted, provides info necessary to replicate experiment, and describes any equipment or testing materials that were used.
Presents the findings, first there is a description in narrative form. Second, the results are described in statistical language. Third, the material is often depicted in tables and graphs.
The researcher reviews the research from various perspectives. Do the results support the hypothesis? What might have been wrong with the methodology, the hypothesis, or both? The researcher may also discuss how the results compare with past research results on the topic. It may also provide suggestions for practical applications of the research.
IRB stands for
institutional review board.
IRB started because of the
Tuskegee Syphilis study.
IRB reviews experiments to make sure they are
The Belmont report
written be the National Commission for the Protection of Human Subjects of Biomedical and Behavioral Research 1979.
The three major principles of the Belmont report are
Beneficence, respect for persons (autonomy), and justice.
What is Beneficence
need for research to maximize benefits and minimize any possible harmful effects of participation.
Respect for Persons (autonomy)
treating them with courtesy and respect and allowing for informed consent. Researchers must be truthful and conduct no deception.
ensuring reasonable, non-exploitative, and well-considered procedures are administered fairly the fair distribution of costs and benefits to potential research participants and equally.
When is an IRB necessary?
All proposed research that involves (1) intervention or interaction with human subjects, (2) the collection of identifiable private data on living individuals and/or (3) data analysis of identifiable private information on living individuals requires review and approval by the IRB prior to the initiation of the research.
any event, situation, behavior, or individual characteristic that varies. A variable must have two or more levels or values.
Variables can be classified into four types. -
situational, response, subject, mediating.
describe characteristics of a situation or environment: the length of words that you read in a book, the spatial density of a classroom, the credibility of a person trying to persuade you.
are the responses or behaviors of individuals such as reaction time, performance on a cognitive test.
individual differences there are characteristics of individuals including gender, intelligence, and personality traits such as extraversion.
are psychological processes that mediate the effects of a situational variable on a particular response. As an example, Darley and Latane found that helping is less likely when there are more bystanders to an emergency.
What does it mean to operationalize a variable?
It is important to know that a variable is an abstract concept that must be translated into concrete forms for observation or manipulation so that it may be measured empirically. The variable must be described in terms of the specific method and operations used by the experimenter to measure or manipulate it. For example the term aggression could mean number of shocks delivered, number or times a child punches inflatable doll, number of times the child fights with others at recess, so on and so forth.
Four Scales of Measurement
nominal, ordinal, interval and ratio measurement.
At the nominal scale, i.e., for a nominal category, one uses labels; for example, rocks can be generally categorized as igneous, sedimentary and metamorphic. For this scale, some valid operations are equivalence and set membership. Nominal measures offer names or labels for certain characteristics. Variables assessed on a nominal scale are called categorical variables.
Rank-ordering data simply puts the data on an ordinal scale. Ordinal measurements describe order, but not relative size or degree of difference between the items measured. In this scale type, the numbers assigned to objects or events represent the rank order (1st, 2nd, 3rd, etc.) of the entities assessed. A Likert Scale is a type of ordinal scale and may also use names with an order such as: "bad", "medium", and "good"; or "very satisfied", "satisfied", "neutral", "unsatisfied", "very unsatisfied."
Quantitative attributes are all measurable on interval scales, as any difference between the levels of an attribute can be multiplied by any real number to exceed or equal another difference. A highly familiar example of interval scale measurement is temperature with the Celsius scale. In this particular scale, the unit of measurement is 1/100 of the temperature difference between the freezing and boiling points of water under a pressure of 1 atmosphere. The "zero point" on an interval scale is arbitrary; and negative values can be used.
Most measurement in the physical sciences and engineering is done on ratio scales. Mass, length, time, plane angle, energy and electric charge are examples of physical measures that are ratio scales.
3 types of Correlations
positive linear relationship, negative linear, and curvilinear
Positive Linear Relationship
increases in the values of one variable are accompanied by an increase in the value of the second variable.
Negative Linear Relationship
Increases in the values of one variable are accompanied by a decrease in the values of the second variable.
increases in the values of one variable are accompanied by both increases and decreases in the values of the other variable.
The experimental method involves direct
manipulation and control of variables. The researcher manipulates the first variable of interest and then observes the response. With this method, the two variables do not merely vary together; one variable is introduced first to see whether it affects the second variable.
all extraneous variables are kept constant. If a variable is held constant, it cannot be responsible for the results of the experiment. In other words, any variable that is kept constant cannot be a confounding variable. Experimental control is accomplished by treating participants in all groups in the experiment identically; the only difference between groups is the manipulated variable.
sometimes it is difficult to keep a variable constant such as any characteristic of the participants. The experimental method eliminates the influence of such variables by randomization. Randomization ensures that the extraneous variable is just as likely to affect one experimental group as it is to affect the other group. The ability to randomly assign research participants to the conditions in the experiment is an important difference between experimental and non-experimental methods.
Some advantages of the non
experimental method are - that it can be applied where the experimental method would be either unethical or impractical. An example would be studies of child-rearing practices.
refers to truth and the accurate representation of information.
refers to the adequacy of the operational definition of variables: does the operational definition of a variable actually reflect the true theoretical meaning of the variable? Many variables are abstract constructs such as social anxiety, speaker credibility, or social loafing. The measure has construct validity if it measures the social anxiety construct and not some other variable such as dominance.
refers to the ability to draw conclusions about causal relationships from our data. A study has high internal validity when strong inferences can be made that one variable caused changes in the other variable.
the extent to which the results can be generalized to other populations and settings. Can the results be replicated with other operational definitions of the variables, with different participants in different settings?
A confounding variable
a variable that varies along with the independent variable; confounding occurs when the effects of the independent variable and an uncontrollable variable are intertwined so you cannot determine which of the variables is responsible for the observed effect.
When the results of an experiment can confidently be attributed to the effect of the independent variable. To achieve, the researcher must design and conduct the experiment so that only the independent variable can be the cause of the results.
Post test only design a researcher must (3)
t- (1) obtain two equivalent groups of participants, (2) introduce the independent variable, and (3) measure the effect of the independent variable on the dependent variable.
Pretest Post test Design
the only difference between the two is that a pretest is given before the experimental manipulation is introduced. This design makes it possible to ascertain that the groups were, in face, equivalent at the beginning of the experiment.
Repeated Measures Design
the same individuals will participate in both conditions. Participants are repeatedly measured on the dependent variable after being in each condition of the experiment.
Advantages of a repeated measures design
Fewer participants are needed because each individual participates in all conditions. Extremely sensitive to finding statistically significant differences between groups. It is much easier to separate the systematic individual differences from the effect of the independent variable.
Disadvantages of a repeated measures design
The major pitfall of repeated measures design is that the different conditions must be presented in a particular sequence.
the order of presenting the treatments affects the dependent variable. There are several types of order effects such as practice effect, fatigue effect, and contrast effect.
with complete counterbalancing, all possible orders of presentation are included in the experiment. Some are assigned to the low/high order and others are assigned to the high/low order. By counterbalancing your design, it is possible to determine the extent to which order is influencing your results. A technique used to control for order effects is called the Latin Squares:
Technique used to control for counter balancing is
- latin squares.
a limited set of orders constructed to ensure that (1) each condition appears at each ordinal position and (2) each condition preceded and follows each condition one time.
sometimes it is necessary to stage events that occur during the experiment in order to manipulate the independent variable successfully.
Sensitivity of a measure
Do you like this person? Yes or No- only 2 choices, not very sensitive. If you used a 7 point scale it would be more sensitive.
any feature of an experiment that might inform participants of the purpose of the study. The concern is that when participants form expectations about the hypothesis of the study, they will do whatever is necessary to confirm the hypothesis. One way to control demand characteristics is to use deception. Another way is to use filler items on your test. Lastly, people could not be aware that an experiment is even taking place.
since experimenters know the purpose and design of the study they develop expectations about how the participants should respond.
There are two potential sources of experimenter bias
1. The experimenter might unintentionally treat participants differently in the various conditions of the study. 2. When experimenters record the behaviors of the participants; there may be subtle differences in the way the experimenter interprets and records the behaviors.
Three Ways to solve Expectancy Problem
1 Experimenters should be well trained and should practice behaving consistently with all participants. 2 Run all conditions simultaneously so that the experimenter's behavior is the same for all participants.3 Procedures are automated. 4 Use experimenters who are unaware of the hypothesis being investigated.
the participant is unaware
both the participant and the experimenter are unaware of who is getting what.
Ttest (t-statistic) -
used to determine if there is a significant difference between groups (used to compare group means)
is an interval/ratio scale
is a two-level categorical variable
Ttest Can be conducted on a-
single one sample, paired matched sample, and independent samples.
(i.e., group mean versus known population mean)
(i.e., pre- and post- treatment means)
(i.e., comparing GPA of males vs. females)
TTESTS ARE LIMITED TO COMPARING THE MEANS OF
ANOVA Analysis of Variance (F statistic)
used to determine if there is a significant difference between groups (used to compare group means, too, like t-test)
The oneway ANOVA is used when you have
a:dependent variable is an interval or ratio scale b:independent variable that is nominal (or ordinal) c:WHEN YOU HAVE MORE THAN 2 GROUPS (means) TO COMPARE (e.g., small vs. medium vs. large font)
ANOVA and ttest are the equivalent when you only have
Independent Variable with only 2 levels.