EXPERIMENTAL PSYCH PRELIMS.pdf

EXPERIMENTAL PSYCHOLOGY PRELIMS 1ST SEMESTER – MS. APPLE PEDROS – 10:00-11:30 Mon/Thurs Homeopathy M1: THE BASICS OF EXPERIMENTATION Using diluted natural substances for AND C...

EXPERIMENTAL PSYCHOLOGY PRELIMS 1ST SEMESTER – MS. APPLE PEDROS – 10:00-11:30 Mon/Thurs Homeopathy M1: THE BASICS OF EXPERIMENTATION Using diluted natural substances for AND CONTROLLING EXTRANEOUS medical treatment. VARIABLES Pyramidology: Theories about pyramids, like extraterrestrials building them or THE SCIENCE OF PSYCHOLOGY healing powers. A WHAT IS SCIENCE? A2 PSYCHOLOGY Science helps us understand the natural The scientific study of behavior and world. mental processes. THREE FEATURES OF SCIENCE: Focuses on actions, thoughts, emotions, Systematic empiricism reasoning, memories, and biological Empirical questions functions. Public knowledge PSYCHOLOGISTS’ GOALS EMPIRICISM Describe phenomena Learning based on observation. Predict future behaviors Scientists learn systematically by Explain causes of behaviors planning, observing, recording, and analyzing. EMPIRICAL QUESTIONS A3 EXPERIMENTAL PSYCHOLOGY These are questions about how the world actually works. Studies sensing, perceiving, learning, and Can be answered by observing thinking. systematically. PUBLIC KNOWLEDGE Science creates public knowledge. A4 SCIENTIFIC METHOD IN PSYCHOLOGY Usually through professional journal articles explaining the research, Approach used by psychologists to methods, and results. acquire knowledge about behavior. STEPS: A1 PSEUDOSCIENCE 1. Identify questions of interest Based on behavior, previous research, Activities/beliefs that claim to be curiosity, creativity, insight. scientific but aren't. 2. Formulate an explanation Based on behavior, previous research, EXAMPLES: curiosity, creativity, insight. Cryptozoology 3. Carry out research Hidden creatures like Bigfoot, Loch Based on behavior, previous research, Ness monster, chupacabra. curiosity, creativity, insight. Pseudoscientific psychotherapies 4. Communicate Findings Past-life regression, rebirthing, Based on behavior, previous research, bioscream therapy. curiosity, creativity, insight. PAGE 1 GIELI ANNE B. DOLERA – BS PSYCH 3 – YOU CAN DO IT!  EXPERIMENTAL PSYCHOLOGY PRELIMS 1ST SEMESTER – MS. APPLE PEDROS – 10:00-11:30 Mon/Thurs B3 OPERATIONAL DEFINITION B VARIABLE Specifies the precise observable Any factor or property that a researcher operations, procedures, and measures, controls or manipulates. measurements the changing quantity or measure of any clearly describes the operations involved factor, trait or condition in manipulating or measuring the It is a logical set of attributes, variables in an experiment characteristics, quantities that can be statements of operating procedures, sets measured or counted. of instructions that tell others how to carry out an experiment B1 INDEPENDENT VARIABLE (IV) B4 DEFINING CONSTRUCTS OPERATIONALLY The condition that is manipulated by an experimenter. Not affected by anything else that HYPOTHETICAL CONSTRUCTS: happens in the experiment. (Constructs that cannot be observed Referred also as treatments, directly) would be too much vague manipulations, interventions and without precise operational definition conditions. thus we infer its occurrence from objective measures EXAMPLES: SELF-ESTEEM AND FACEBOOK PROFILE OF Task Variables STUDENTS Illumination (bright/dim) Typeface size (large/small) Self-esteem Noise level (loud/soft) describes a person’s overall subjective Subject Variables emotional evaluation to oneself Characteristics of the subjects through the Rosenberg self-esteem themselves (age, personality scale. If the respondent has a higher characteristics, gender and etc.) score it will indicate a higher self-esteem and low self-esteem for lower score B2 Dependent Variable (DV) Self-presentation of Facebook Profile Refers to the characteristics of the facebook user presented in his/her The variable that is measured and is facebook account including the post expected to change as a result of updated and photos/videos shared as changes caused by the experimenter’s measured by the Self –Presentation manipulation of the independent scale variable. Referred also as measure, outcomes, results and effects PAGE GIELI ANNE B. DOLERA – BS PSYCH 3 – YOU CAN DO IT! 2  EXPERIMENTAL PSYCHOLOGY PRELIMS 1ST SEMESTER – MS. APPLE PEDROS – 10:00-11:30 Mon/Thurs B5 TYPE OF SCALES / LEVELS OF MEASUREMENT 3. INTERVAL SCALE categories have different names and are organized sequentially and show MEASUREMENT the degree of difference Application of rules for assigning No absolute zero value (only uses zero numbers to objects. The rules are the as a reference point but does not specific procedures to transform qualities indicate absence of the variable being of attributes into numbers. measured. 1. NOMINAL SCALE INTERVAL SCALE EXAMPLE simply represents the qualitative or categorical differences of the variables IQ measured. No such thing as zero IQ just for labeling purposes, nothing more Test Scores A zero score does not mean zero knowledge or the student did not NOMINAL SCALE EXAMPLE learn anything NOMINAL SCALE CATEGORY ºC and ºF scales GENDER Male or Female Have zero value dut does not mean there is no temperature at all. RESPONSES Yes or No RACE Race Asian, 4. RATIO SCALE American, African, et., categories have different names and are organized sequentially and show the degree of difference and can hold 2. ORDINAL SCALE an absolute zero value: categories have different names and The existence of an absolute zero point are organized sequentially. means that we can measure the Gives order or rank to subjects or items absolute amount of the variable. We can determine the difference between categories and the direction RATIO SCALE EXAMPLE of the difference but not the magnitude of the difference. WEIGHT INCOME ORDINAL SCALE EXAMPLE Socioeconomic status AGE Upper, middle, and lower Contest Winners 1st place, 2nd place, etc. T-shirt sizes Small, medium, large, extra large PAGE GIELI ANNE B. DOLERA – BS PSYCH 3 – YOU CAN DO IT! 3  EXPERIMENTAL PSYCHOLOGY PRELIMS 1ST SEMESTER – MS. APPLE PEDROS – 10:00-11:30 Mon/Thurs EVALUATING OPERATIONAL DEFINITIONS A. Face Validity The condition in which a measuring instrument intuitively seems to C1 RELIABILITY measure what it is supposed to measure. B. Content Validity CONSISTENCY AND DEPENDABILITY Depends whether we are taking a fair If we apply them in more than one sample of the variable we intent to experiment, they ought to work in similar measure. ways each time C. Predictive Validity Do the procedures yield information C2 PROCEDURES FOR CHECKING THE that enables us to predict future RELIABILITY MEASUREMENT behavior or performance? TECHNIQUES D. Concurrent Validity Compares scores on the measuring instrument with an outside criterion A. Interrater Reliability E. Construct Validity To have many different observers take Deals with the transition from theory measurements of the same responses to research application; when several B. Test-Retest Reliability measures fit sensibly together and Comparing scores of people who have converge on (and can be explained by) been measured twice with the same an underlying psychological concept. instrument C. Interitem Reliability The extent to which different parts of a C4 EVALUATING THE EXPERIMENT questionnaire, test, or other instruments designed to assess the INTERNAL VALIDITY same variable attain consistent results. Able to state a causal relationship between antecedent conditions and the 2 BASIC APPROACHES subsequent observed behavior. When we set up an experiment, we plan 1. Internal Consistency procedures to measure the effects of Most often used when a researcher has various treatment levels. The goals of an created a multiple-item questionnaire experiment are to assess the impact of to measure a single construct like the independent variable. intelligence, need for achievement, or An experiment is internally valid if we anxiety. can be sure that the changes in 2. Split-half reliability behavior is observed across the Involves splitting the test into two treatment conditions of the halves and computing a coefficient of experiment were actually caused by reliability between the two halves. the independent variable C3 VALIDITY Principle of actually studying the variables that we intend to study. PAGE GIELI ANNE B. DOLERA – BS PSYCH 3 – YOU CAN DO IT! 4  EXPERIMENTAL PSYCHOLOGY PRELIMS 1ST SEMESTER – MS. APPLE PEDROS – 10:00-11:30 Mon/Thurs CLASSICAL THREATS OF INTERNAL VALIDITY INSTRUMENT Happens when Ex: Imagine ATION some feature of that your HISTORY Possible Ex: societal the dependent confound in (e.g., World measuring measure was research that War II instrument itself the length of inadvertently and the changes during a line drawn takes place Vietnam the experiment by the between War) unique subjects. To measurements to the time measure because of of test it, you use historical may have a the only ruler changes in the significant you have— a participant. effect on the rubber ruler. behavior of Unknown to interest (e.g., you, it expressed stretches a attitudes bit every about the time you use war)events it. Each consecutive MATURATIO Any internal Ex: Boredom, measuremen N (physical or fatigue, etc., ts tend to psychological) it must underestima changes in also be te tend to subjects that considered underestima might have for te line length affected scores longitudinal on the studies. STATISTICAL The tendency for dependent REGRESSION extreme measure (not measures on just “maturing” some variable to in the way be we typically closer to the think of it, group mean although it when could be) remeasured, owing to TESTING Refers to effects unreliability of on the measurement dependent variable SELECTION Whenever the produced by a researcher does previous not administration assign subjects of the same test randomly to the or different other measuring conditions of an instrument. experiment, the PAGE GIELI ANNE B. DOLERA – BS PSYCH 3 – YOU CAN DO IT! 5  EXPERIMENTAL PSYCHOLOGY PRELIMS 1ST SEMESTER – MS. APPLE PEDROS – 10:00-11:30 Mon/Thurs Needs to be detailed enough that subjects another researcher could read it and in one replicate what you did. treatment Divided into labeled subsections such as condition may participants, materials, and procedure, begin the but a great deal of flexibility in labeling experiment with can accommodate specific experiments. different characteristics than those of A. Participants (or subjects) section the subjects in Tells how many subjects were another How subjects were selected condition Who they were B. Apparatus Section SUBJECT When subjects Any equipment used to test the MORTALITY dropped subjects. out of one Include details such as the model experimental number of a computer or the size of a condition than conditioning chamber. the other. Referred to as the materials section when questionnaires, written, or SELECTION Family of videotaped sketches, and other similar INTERACTIO threats; a means are used to test subjects. NS selection threat If they are long, special materials may can combine be placed in an appendix section. with another Usually set in a smaller type. threat to C. Procedure Section form a selection Explains what happened to the interaction. If subjects and includes instructions (for subjects were human subjects), statistical design not randomly features, and so forth. assigned If an uncommon technique was used- in to groups, any that is, one that cannot be looked up one of the other directly in an advanced statistics text threats may and cited-an extra design subsection is have affected often included. Sometimes even a some standard statistical technique is experimental described in a design subsection. groups but not others C5 CONTROLLING EXTRANEOUS VARIABLES C4 PLANNING THE METHOD SECTION PHYSICAL VARIABLES METHOD SECTION OF THE RESEARCH PAPER Aspects of the testing conditions that The place to describe what you did in need to be controlled your experiment (who, what, when, and Example: noise, distractions such as how). extreme temperature in the testing room, weather, lighting, etc., PAGE GIELI ANNE B. DOLERA – BS PSYCH 3 – YOU CAN DO IT! 6  EXPERIMENTAL PSYCHOLOGY PRELIMS 1ST SEMESTER – MS. APPLE PEDROS – 10:00-11:30 Mon/Thurs THREE GENERAL TECHNIQUES FOR Single-Blind Experiments CONTROLLING PHYSICAL VARIABLES: an experiment in which subjects do not know which treatment they are getting. 1. Elimination Another alternative for controlling the Take out the extraneous variable possibility that subjects may guess the Ex: Hanging a sign on the door, “Do not experimental hypothesis—we could use disturb. Experiment in progress.” a cover story; a plausible explanation for 2. Constancy of conditions the procedures used in the study. It is Keep all aspects of the treatment told to disguise the actual research conditions as nearly similar as possible. hypothesis so that subjects will not guess If we cannot eliminate an extraneous what it is. variable, we try to make sure that it Experimenter Bias stays the same in all treatment refers not only to the conscious cases of conditions. fraud but also to the much more subtle 3. Balancing influences that experimenters may Distributing the effects of an unknowingly exert on the outcome of extraneous variable across the different their research treatment conditions of the Example: Errors in recording the data, experiment misreading a scale or scoring an item Example: incorrectly, etc., Coincidentally, Rosenthal (1987) reported that researchers are more likely to make errors that favor the hypothesis Rosenthal Effect (also called Pygmalion Effect) When experimenter treats subjects differently depending on what they expect from them. C7 CONTROLLING EXPERIMENTER BIAS SOCIAL VARIABLES Double-Blind Experiments Qualities of the relationships between an experimental technique in which subjects and experimenters that can neither the subject nor the experimenter influence results knows which subjects are in which Demand Characteristics. Aspects of the treatment conditions. experimental situation that demand that PERSONALITY VARIABLES people behave in a particular way. An Experimenters. Characteristics, or experimenter generally wants personality variables, that an participants to be as naïve as possible. experimenter brings to the experimental They should understand the nature and settings can be important. purpose of the experiment but not the Example: An experimenter who is warm exact hypothesis and friendly can elicit very different responses from subjects that can one C6 CONTROLLING DEMAND who is cold and aloof. CHARACTERISTICS CONTEXT VARIABLES Those that come about from procedures PAGE GIELI ANNE B. DOLERA – BS PSYCH 3 – YOU CAN DO IT! 7  EXPERIMENTAL PSYCHOLOGY PRELIMS 1ST SEMESTER – MS. APPLE PEDROS – 10:00-11:30 Mon/Thurs created by the environment, or context, of the Between-Subjects Design research setting; includes subject recruitment, Definition: Participants are divided selection, and assignment procedures, as well as into different groups, and each group typical problems encountered in research on a experiences only one condition or level university population. of the independent variable. Goal: To determine whether 2 KINDS OF CONTEXT VARIABLE differences exist between two or more 1. When the Subjects Select the treatment conditions. Experiment Advantages of Between-Subjects Design A biased sample greatly reduces your No Contamination: No risk of ability to generalize your findings; the treatment carryover effects since experiment will have less external participants only experience one validity treatment. 2. When the Experimenter Selects the Reduces Practice/Fatigue Effects: Subjects Participants are not subjected to If you do not select your subjects repeated tasks, reducing fatigue or randomly, your sample will be biased, learning effects. resulting in less generalizability. Feasibility: Easier to implement when exposing participants to all conditions isn’t possible or ethical. M2: BASIC BETWEEN-SUBJECTS Generalizability: More representative DESIGNS of the broader population since prior experience with other conditions is excluded. A EXPERIMENTAL DESIGNS Random Assignment: Helps control for individual differences, isolating the effect of the independent variable. Methods used to allocate participants to Disadvantages of Between-Subjects Design different conditions and collect data at Individual Differences: Inherent specific times. differences between participants may affect results, leading to confounding A1 TYPES OF EXPERIMENTAL DESIGNS variables. Time and Resources: Conducting experiments with separate groups BETWEEN-SUBJECTS DESIGN requires more time and resources. Attrition: Participants might drop out WITHIN-SUBJECTS DESIGN or not comply with instructions, affecting the reliability of the results. LONGITUDINAL DESIGN Sample Size: A larger sample size is needed to ensure findings are reliable, CROSS-SECTIONAL DESIGN increasing the complexity of the study. Variability: Results might vary more FACTORIAL DESIGN between individuals, making it harder to detect small effects. SMALL-N DESIGN PAGE GIELI ANNE B. DOLERA – BS PSYCH 3 – YOU CAN DO IT! 8  EXPERIMENTAL PSYCHOLOGY PRELIMS 1ST SEMESTER – MS. APPLE PEDROS – 10:00-11:30 Mon/Thurs If individuals in the population are all A2 THINGS TO CONSIDER very similar to one another on the dependent variable, small samples are adequate. SELECTING AND RECRUITING SUBJECTS When individuals are likely to be quite different, larger samples are needed. A general rule, it is advisable to have at A3 WHAT IS YOUR SAMPLING least 20 subjects in each treatment PROCEDURE? group; however, most researchers are more comfortable if there are 30 subjects in each group. Smaller numbers make it SIMPLE RANDOM SAMPLING very difficult to detect an effect of the independent variable CLUSTER SAMPLING SYSTEMATIC SAMPLING A5 APPLICATION OF BETWEEN-SUBJECTS DESIGNS CONVENIENCE SAMPLING TWO-GROUP DESIGN PURPOSIVE SAMPLING The simplest experiments are those in which there is only one independent QUOTA SAMPLING variable. TWO VARIANTS OF TWO GROUP DESIGN SNOWBALL SAMPLING A. TWO-INDEPENDENT-GROUP DESIGN JUDGEMENTAL SAMPLING B. TWO-matched-group design MULTI-STAGE SAMPLING TWO-INDEPENDENT-GROUP DESIGN ROSNOW AND ROSENTHAL (1976) Subjects are placed in each of two SUGGESTIONS treatment conditions through random Make your appeal interesting, assignment. nonthreatening, and meaningful. Get Random assignment means that every someone known to the subject to ask subject has an equal chance of being them, preferably a woman of High Status. placed in any of the treatment Emphasize the responsiblity of people to conditions. When we use an aid in research that they it can help independent group design, we use the others and lots of people do it. Pay them same unbiased procedures for assigning if you can and give token gifts for taking subjects to groups that are used in time to hear you out. Assess whether you random selection of subjects. should ask for volunteers or privately. Try to put yourself in the other’s place and ask what make you most likely to say yes. A4 HOW MANY SUBJECTS ARE NEEDED? THINGS TO CONSIDER! PAGE GIELI ANNE B. DOLERA – BS PSYCH 3 – YOU CAN DO IT! 9  EXPERIMENTAL PSYCHOLOGY PRELIMS 1ST SEMESTER – MS. APPLE PEDROS – 10:00-11:30 Mon/Thurs - To assess the impact of the independent or the Sedentary Group. This helps variable, we must have at least two different ensure that any individual differences treatment conditions so that we can compare are evenly distributed across the the effect of different values of the groups. independent variable. Treatment Phase: Experimental Condition Exercise Group: Participants in this We apply a particular value of our group engage in regular exercise for a independent variable to the subjects specified period (e.g., 30 minutes of and measure the dependent variable moderateintensity exercise, five days a The subjects in an experimental week) condition are called an experimental / Sedentary Group: Participants in this treatment group group are instructed to maintain their Control Condition current sedentary lifestyle without Used to determine the value of the engaging in any structured exercise. dependent variable without an experimental manipulation of the EXAMPLE: independent variable. In this condition, exactly the same procedures are Measurement of Blood Pressure: carried out in the experimental Both groups have their blood pressure condition, except for the experimental measured before the study (baseline manipulation. measurement) and after the treatment phase EXAMPLE: Data Analysis: Research Question: The average reduction in blood Does regular exercise lead to a pressure for each group is calculated significant reduction in blood pressure and compared. compared to a sedentary lifestyle? Expected Outcome: Experimental Setup: If regular exercise is effective in Independent Variable (IV): Exercise reducing blood pressure, it is (Regular Exercise vs. Sedentary anticipated that the Exercise Group Lifestyle) will show a greater reduction in blood Dependent Variable (DV): Blood pressure compared to the Sedentary Pressure Group. Experimental Process: TWO-MATCHED-GROUP DESIGN Participant Recruitment: Participants There are also two groups subjects, but are recruited for the study. They may the researcher assigns them to groups by be individuals with high blood matching or equating them on a pressure who are interested in characteristic that will probably affect the participating in the study. dependent variable. The researcher forms the groups in such a way they are EXAMPLE: sure to be comparable on an extraneous variable that might otherwise produce Random Assignment: confounding. Participants are randomly assigned to one of two groups: the Exercise Group PAGE GIELI ANNE B. DOLERA – BS PSYCH 3 – YOU CAN DO IT! 10  EXPERIMENTAL PSYCHOLOGY PRELIMS 1ST SEMESTER – MS. APPLE PEDROS – 10:00-11:30 Mon/Thurs WE CAN DO THESE IN THREE WAYS: ATTRIBUTED TO THE TEACHING METHODS 1. Precision matching. We insist that the RATHER THAN PRE-EXISTING DIFFERENCES members of the matched pairs have BETWEEN THE GROUPS identical scores 2. Range matching. We require that the members of a pair fall within a previously A6 WHEN TO USE TWO MATCHED specified range of scores. GROUPS 3. Rank-ordered matching. Subjects are simply rank ordered by their scores on the matching variable, and subjects with ADVANTAGE: adjacent scores then become a matched By matching on a variable that is likely to pair. have a strong effect on the dependent variable, we can eliminate one possible EXAMPLE: source of confounding. We do not need to assume that our treatment groups are Research Question: comparable through matching. Does a new teaching method (Method A) improve math scores compared to the traditional teaching method A7 MULTI-GROUP DESIGN (Method B)? MULTIPLE GROUP DESIGN: Experimental Design: There are more than two groups of Participant Selection: Researchers subjects and each group is run through a select a sample of 60 students from a different treatment condition. One of the local high school. The students are treatment conditions may be a control assessed on their previous math scores condition in which subjects receive the and their learning styles (e.g., visual, zero value of the independent variable. auditory, kinesthetic). The most commonly used Matching: The researchers pair multiple-groups design is the students based on their previous math multiple-independent groups design in scores and learning styles. For which the subjects are assigned to the example, Student 1 (Method A) will be different treatment conditions at matched with Student 2 (Method B), random. Student 3 (Method A) with Student 4 (Method B), and so on. ASSIGNING SUBJECTS: Random Assignment: Within each One way if to use random number matched pair, the researchers table: We could first pick a starting point randomly assign one student to the in the random number table in an experimental group (Method A) and unbiased way, for instance, by closing our the other to the control group (Method eyes and pointing. B). CHOOSING TREATMENTS: IN THIS EXAMPLE, THE MATCHING PROCESS Frequently, experimenter will select HELPS TO ENSURE THAT BOTH GROUPS ARE treatment conditions that are COMPARABLE IN TERMS OF THEIR INITIAL proportional, such as drug doses of, say, MATH ABILITY AND LEARNING STYLES. THIS 5, 10, and 15 milligrams of exercise periods MAKES IT MORE LIKELY THAT ANY of 15, 30, and 45 minutes. DIFFERENCES IN MATH SCORES OBSERVED AFTER THE INTERVENTION CAN BE PAGE GIELI ANNE B. DOLERA – BS PSYCH 3 – YOU CAN DO IT! 11  EXPERIMENTAL PSYCHOLOGY PRELIMS 1ST SEMESTER – MS. APPLE PEDROS – 10:00-11:30 Mon/Thurs PRACTICAL LIMITS: structure and the required treatment The multiple-groups procedure assumes conditions. that treatment groups are formed by random assignment. Thus, there will be B1 DESCRIBING THE DESIGN as many different treatment groups in the experiment as there are levels of the independent variable. If you have 5 levels There are special techniques used to of the iv, you will need 5 groups of describe factorial designs in a way that subjects. It might be difficult to find corresponds to the statistical analyses enough subjects to make this design that are needed to analyze them. feasible A two-factor experiment has two IVs and each IV has two levels. This design can be written in shorthand form as a 2x2 M3: BETWEEN-SUBJECTS FACTORIAL (read as “two by two”) factorial design. DESIGN This shorthand notation tells us several things about the experiment it describes. First, the numbers tell us the number of A FACTORIAL DESIGN factors involved. Here there are two numbers (2 and 2); each number refers Designs in which two or more to a different factor. Hence, this independent variables (IVs) are studied experiment has two factors. The simultaneously. numerical value of each number tells is IVs are called factors, and each factor has how many levels each factor has. If we two or more values or levels. have a 2x2 design, we automatically A two-factor experiment (the simplest know that each of the experiment’s two factorial design) has two factors only. factors has two levels. We also know The results we get from a factorial design that the experiment has four different experiment give us two kinds of conditions (the product of 2x2). information: MAIN EFFECTS: B2 FACTOR LABELING METHODS Effects of each IV in the experiment. The action of a single IV in an experiment. The names of each factor are placed in Measures how changes in one IV affect parentheses following the numerical subjects' behaviors. notations. A main effect is simply a change in behavior associated with a change in the Example 1: 2x2 (Type of Name X Length value of a single IV. of Name) between-subjects factorial INTERACTION: design Determines how the influence of one IV Example 2: 2(type of name) x 2 (length of affects the influence of another. name) between- subjects factorial design is present if the effect of one IV changes across the levels of another IV. These methods also can be expanded to include the factor levels. You can include the factor name (capitalized) and its B LAYING OUT A FACTORIAL DESIGN levels (lowercase), separated by a colon, or you can simply use the levels Visualize the design graphically using a (lowercase). design matrix to understand the PAGE GIELI ANNE B. DOLERA – BS PSYCH 3 – YOU CAN DO IT! 12  EXPERIMENTAL PSYCHOLOGY PRELIMS 1ST SEMESTER – MS. APPLE PEDROS – 10:00-11:30 Mon/Thurs Example 3: 2x2 (Type of Name: given, nickname x Length of Name: short, long) between-subjects factorial design Example 4: 2 (given name of nickname) x 2 (short or long name) If an experiment is referred to as a “2x3x2 factorial design,” we immediately know several things about it. Three digits mentioned (2,3, and 2), so we know this experiment involves 3 factors. The numerical value of each digit tells us the number of levels of each factor. We know that the first factor has two levels, the second has three, and the third has two. We also know that this experiment has 12 separate conditions (the product of 2x3x2). PAGE GIELI ANNE B. DOLERA – BS PSYCH 3 – YOU CAN DO IT! 13