Chapter 5 & 6: Correlational and Quasi-Experimental Design PDF

ALTERNATIVE TO EXPERIMENTATION: CORRELATIONAL AND QUASI-EXPERIMENTAL DESIGN CHAPTER 5 & 6 PSY 213/L – EXPERIMENTAL PSYCHOLOGY (1296) UM Tagum College Submitted to:...

ALTERNATIVE TO EXPERIMENTATION: CORRELATIONAL AND QUASI-EXPERIMENTAL DESIGN CHAPTER 5 & 6 PSY 213/L – EXPERIMENTAL PSYCHOLOGY (1296) UM Tagum College Submitted to: Prof. Lloyd Sajol, MPsy Submitted by: Baring, Cielo Marie Liba, Bambie Shayme Ococa, Janica Malaque Paticanon, Marjorie Peñaflor, Ericka Louisse Aug 2024  CHAPTER 5: Alternative to Experimentation: Correlational and Quasi-Experimental Design TERMS DEFINITION OF TERMS REMARKS 5.1 Correlational Designs  Correlational designs are used to establish relationships among preexisting behaviors and can be used to predict one set of behaviors from others (such as predicting your college grades from scores on your entrance exam). Correlational Designs  Correlational designs can show relationships between sets of antecedent conditions and behavioral effects.  They are neither manipulated nor controlled by the researchers. 5.1.1 Correlation Correlation, or degree of relationship, between two traits, behavior, or events, represented by r. Correlation is common in nonexperimental studies that is discussed as a research method Correlation in its own right. Correlation can be used with both laboratory and field data. Correlation is really a statistical technique for summarizing data that could be used in studies falling in any portion of our graphic scheme.  A correlation study is one that is designed to determine the correlation, or degree of relationship, between two traits, behaviors, or events.  In correlational study, the researchers measure events without attempting to alter Correlation Study the antecedent conditions in any way; the researcher is simply asking how well the measures go together  Researchers often use correlational studies to explore behaviors that are not yet well understood. 5.1.1.1 Kinds of Correlational Study  Variable is any observable behavior, characteristics, or event that can vary or have different values. Variable  Examining cause-and-effect correlations, researchers frequently measure or alter independent and dependent variables.  Aiding us in the discovery of important influences on behavior. Heuristic Value  A technique for conducting scientific research using procedures and methods meant to make discoveries.   May serve as the basis for new experimental hypothesis. Correlation data  To determine which variables are connected, researchers use correlations.  Relationship between pairs of scores from each subject. Simple Correlation  The degree of relationships, or correlation, between the two measures would then be evaluated through statistical procedures.  Scatterplot is a visual representation of the scores belonging to each subject in the study.  A graph of data from a correlational study, created by plotting pairs of scores from each Scatterplot subject; the value of one variable is plotted on the X (horizontal) axis and the other variable on the Y (vertical) axis.  Scatterplots are often the researcher’s first step toward analyzing correlational data. The most commonly used procedure for calculating simple correlations. Pearson Product Moment Correlation Coefficient (r) The Pearson r is used when interval or ratio scale data are collected.  Correlation coefficient are computed when the data from both measurements are ordinal, nominal, or one of the two is nominal. Correlation Coefficient  Correlation coefficient can be depicted on a number line going from -1.00 to +1.00, like the one below: -1.00 ---------- -.50 ---------- 0 ---------- +.50 ---------- +1.00  The line of best fit; represents the equation that best describes the mathematical relationship between two variables measured in a correlational study. Regression Line  They illustrate the mathematical equation that best describes the linear relationship between two measured scores.  The direction of the line corresponds to the direction of the relationship.  The relationship between two measures such that an increase in the value of one associated with an increase in the value of the other. Positive Correlation  It is also called direct relationship.  Example: Positive correlation (that is, r is positive) between vocabulary and TV viewing time; the more a person watches television, the larger his or her vocabulary. The relationship existing between two variables such that an increase in one is associated with a decrease in the other. It is also called an inverse relationship. Negative Correlation Example: Negative correlation between vocabulary and TV viewing time (that is, r is negative). This would mean that the more a person watches television, the smaller his or vocabulary would be. Coefficient of Determination  The coefficient of determination estimates the amount of variability in scores on one (r2) variable that can be explained by the other variable—an estimate of the strength of the relationship between them.   Also it is easy to compute once we have calculated the ꭇ. 5.1.2 Linear Regression Analysis  When two behaviors are strongly related, the researcher can estimate a score on one of the measured behaviors from a score from other. Linear Regression Analysis  For example, that time spent watching TV and scores on a vocabulary test is correlated, we would substitute someone’s viewing time into the equation for the regression line.  Regression equation is a formula for a straight line that best describes the relationship between the two variables.  It is the equation for a straight line that has both a slope (the direction of the line) and an Regression Equation intercept (the value on the Y, or vertical, axis when X = 0).  Solving the regression equation would give us estimate of what the person’s performance should be on the vocabulary test. 5.1.3 Multiple Correlation and Multiple Regression Statistical intercorrelations among three or more behaviors, represented by R. Conceptually, R is quite similar to r, but R allows us to use information provided by two or Multiple Correlation more measured behaviors to predict another measured behavior when we have that information available.  This analysis allows the statistical influence of one measured variable to be held constant while computing the correlation between the other two. Partial Correlation  If age is an important third variable that is largely responsible for both to increased television viewing and increased vocabulary, statistically controlling for the contribution of age should greatly decrease the correlation between television viewing time and vocabulary.  A correlation-based technique (from multiple correlation) that uses regression Multiple Regression equation to predict the score on one behavior from scores on the other related Analysis behavior.  Multiple regression analyses are very common in the literature. 5.1.4 Factor Analysis  A common correlation procedure that is used when individuals are measured on a large number of items.  Factor analysis allows us to see the degree of relationship among many traits or behaviors Factor Analysis at the same time.  Factor analysis can identify the important dimension underlying a large number of responses. 5.2 Causal Modeling  It is one inherent drawback of all correlational design, the problem of the direction of cause and effect. Causal Analysis  Researchers have tools for causal modeling in correlation-based designs, such as path analysis and cross-lagged panel designs.  5.2.1 Path Analysis  Is an important correlation-based research method that can be used when subjects are measured on several related behaviors.  Is another descriptive method, but it generates important information for prediction and can Path Analysis generate experimental hypothesis.  Uses beta weights to construct path models, outlining possible causal sequences for the related behaviors. 5.2.1.1 Path Model 1. Empathy manipulation 2. Self-report Empathy 3. Attitude toward the group 4. Helping the group 5.2.2 Cross-Lagged Panel Design  This design uses relationships measured over time to suggest the casual path. Cross-Lagged Panel Design  Measures the same pair of variables at two different points in time; looks at patterns of correlations across time for possible direction of cause and effect. 5.3 Quasi-Experimental Designs  Quasi-experimental designs can seem like a real experiment, but they lack one or more of its essential elements, such as manipulation of antecedents or random Quasi-Experimental assignment to treatment conditions. Designs  Quasi-experiments can be used to explore the effects of different treatments on preexisting groups of subjects or to investigate the same kinds of naturally occurring events, characteristics, and behaviors that we measure in correlational studies. 5.3.1 Ex Post Facto Studies A study in which a researcher systematically examines the effects of pre-existing subject characteristics (often called subject variables) by forming groups based on Ex Post Facto Studies these naturally occurring differences between subjects. Ex post facto means “after the fact”. 5.3.2 Nonequivalent Groups Design  A design in which the researcher compares the effects of different treatment Nonequivalent Groups conditions on preexisting groups of participants. Design  The researcher cannot exert control over who gets each treatment because random assignment is not possible. 5.3.3 Longitudinal Design  A method in which the same group of subjects followed and measured at different Longitudinal Design points in time; a method that looks for changes across time.   Longitudinal designs are used in all area of psychology, but they are particularly important for psychologists studying human (and animal) growth and development.  Longitudinal studies are time consuming and hard to conduct. 5.3.4 Cross-Sectional Studies  A method in which different groups of subjects who are at different stages are measured at a single point in time; a method that looks for time-related changes. Cross-sectional Studies  A cross-sectional study will require more subjects; the more groups to be compared, the more subjects needed. 5.3.5 Pretest/Posttest Design  A research design used to assess whether the occurrence of an events alters behavior; scores from measurements made before and after the event are compared.  This design may be used to assess the effects of naturally occurring events (like approval ratings before after a presidential speech) when a true experiment is not possible. Pretest/Posttest  The pretest/posttest design can be used in the laboratory to measure the effect of a treatment presented to subjects by researchers, but the design has a number of problems that reduce its internal validity.  Sometimes, a pretest/posttest design is used in circumstances where the time between the pretest and the posttest is short, as in single laboratory session. 5.3.5.1 Number of Comparison Groups 1. A nonequivalent group- a group that took both the pretest and posttest but was not exposed to the treatment 2. A group that received the treatment and took only the posttest. 3. A posttest-only group.  CHAPTER 6: Formulating The Hypothesis TERMS DEFINITION OF TERMS REMARKS 6.0 Formulating the Hypothesis 6.1 The Characteristics of an Experimental Hypothesis  A tentative explanation of an event or behavior The Characteristics of an  A statement that explains the effects of specified antecedent conditions on a measured Experimental Hypothesis behavior. 6.1.1 Synthetic Statements  Are those that can be either true or false. Synthetic Statements  That can be supported or contradicted. 5.2.1 Path Analysis  Is an important correlation -based research method that can be used when subjects are measured on several related behaviors. Path Analysis  beta weights to construct path models, outlining possible causal sequences for the related behaviors Is another descriptive method, but it generates important information for Uses  prediction and can generate experimental hypotheses.. 6.1.1.1 There are two categories 1. One that is always true Analytic Statement 2. When starting a hypothesis, we want to be concise enough to be proven wrong.  1. A statement of elements that oppose each other. Contradictory Statement 2. The statements are always false. 6.1.2 Testable Statements  It means for manipulating antecedent conditions and measuring the Testable Statements resulting behavior must exist.  Some of the Hypothesis are currently of no scientific use because they do not meet the criterion. 6.1.3 Falsifiable Statements  Statements of research hypotheses must be disprovable by the research findings. Falsifiable Statements  Hypotheses need to be worded so that failures to find the predicted effect must be considered evidence that the hypothesis is indeed false 6.1.4 Parsimonious Statements  Means that the simplest explanation is preferred. Parsimonious  Sometimes called “Occam's razor” Statements 6.1.5 Fruitful Statements  That is it leads to new studies. Fruitful Statements  A hypothesis is fruitful when we can think of new studies that will become important if the hypothesis is supported. 6.2 Inductive Model  Formulating a hypothesis, the process of reasoning from specific cases to more general principles. Inductive Model  You observed several specific instances of behavior and used these instances to form a general principle to explain the behavior.  Induction is the basic tool of theory building.  6.3 Deductive Model  The deductive model of formulating a hypothesis is the reverse of the inductive model. Deductive Model  Deduction is the process of reasoning from general principles to make predictions about specific instances. 6.4 Combining Induction and Deduction  Both induction and deduction are important in research, and both are useful in formulating hypotheses for study. Combining Induction and  Through induction, we devise gen- eral principles and theories that can be used to Deduction organize, explain, and predict behavior until more satisfactory principles are found.  Through deduction we rigorously test the implications of those theories. 6.5 Building on Prior Research  The most useful way of finding hypotheses is by working from research that has already been done.  Sometimes, nonexperimental studies can suggest cause-and-effect explanations that can be translated into experimental hypotheses Building on Prior Research  Prior experimental research is an excellent source of hypotheses.  Regardless of where an experimental hypothesis originates, reviewing the literature is still a necessary component of report writing.  An important goal of report writing is to integrate your findings into existing facts. A good literature review will also help you avoid duplicating someone else's work when replication is not what you had in mind 6.6 Serendipity and the Windfall Hypothesis   is knack of finding things that are not being sought. Serendipity  through serendipity made it to the physical science as well as to psychology.  According to Pavlov, serendipity can be useful in generating new hypothesis when we open new opportunities. 6.7 Intuition  It may be defined as knowing Without reasoning but it is also closest to the phenomenolgy. Intuition  It guided on what we choose to study.  According to Herbert Simon, Intuition is more accurate when it comes from the expert.  The more we know the topic the bitter the intuition hypothesis like to be. 6.7 When All Else Fails  As Russell said, there a no rules that can be used to generate hypothesis. When All Else Fails  Set realistic goals for yourself. Work from hypotheses that can be tested in the time frame you have available.  If you feel completely lost, here are some suggestions that have helped other students. 6.7.1 Suggestions that may helped other students 1. Pick a psychology journal from your library’s shelves and just read through an issue 2. Observation: Some very good hypotheses come from observing how people behave in public places 3. Finally, if all else fails, turn your attention to a real-world problem and try to figure out what causes it. 6.8 Searching the Research Literature  6.8.1 Getting Started  Once you have decided on a hypothesis , you will want to become familiar with other published studies in your topic area. Getting Started  Conducting a thorough literature search is an important part of conducting research, and it’s necessary for writing a research report.  Conducting a literature search is bound to seem daunting at first because there a simply so many sources available. 6.8.1.1 Meta-Analysis  A good source of information that conducted on your topic, which can be found in either journals or edited volumes. Meta-Analysis  A statistical reviewing procedure that uses data from many similar studies to summarize research findings about individual topics.  Has the added benefit of quantifying past findings. 6.8.2 Writing the Report  Public research reports from psychological journals will inform the bulk of the reading you are expected to do as background for writing a research report. Writing the Report  Another goal of a research report is to integrate your experiment into the existing body of knowledge:  To show how the result of your research advance knowledge, increase of generalizability of known effects, or contradicts past findings. 6.8.3 Findings The Article You Need  Fortunately, many library and Web-based aids can help you find the journal articles you need.  The primary resource for psychologist is PsycINFO, an online database published by the American Psychological Findings The Article Association. You Need  Another good source of journal articles (after you have identified key people) is the Social Sciences Citation Index located in the print in your university library or online through the Web of Science. PRACTICE SET 1 (MULTIPLE CHOICE QUESTIONNAIRE) 4. It is the relationship existing between two variables such that an increase in one is associated with a decrease in the other. 1. It is a degree of relationship, between two traits, behavior, or events, a. Positive Correlation represented by r. b. Negative Correlation a. Correlation c. All of the above. b. Correlational Studies d. None of the above. c. Correlation Coefficient d. Simple Correlation 5. In multiple correlation, statistical intercorrelations among three or more behaviors, represented by? What letter? 2. It is the one that is designed to determine the correlation, or degree of a. X b. R relationship, between two traits, behavior, or events. c. S a. Correlation d. Y b. Correlation Study c. Correlation Coefficient 6. It is a relationship between pairs of scores from each subject. d. Simple correlation a. Correlation b. Correlation Coefficient c. Correlation Study 3. It is the relationship between two measures such that an increase in the d. Simple Correlation value of one associated with an increase in the value of the other. a. Positive Correlation b. Negative Correlation 7. This analysis allows the statistical influence of one measured variable to c. All of the above. be held constant while computing the correlation between the other two. a. Multiple Correlation d. None of the above. b. Partial Correlation c. Correlation  d. Factor Analysis 13. It is also called an inverse relationship. a. Positive Correlation 8. It is a visual representation of the scores belonging to each subject in the b. Negative Correlation study. c. Correlation a. Scatterplot d. Both a and b. b. Regression Line c. Correlation Coefficient 14. A correlation-based technique (from multiple correlation) that uses d. None of the above. regression equation to predict the score on one behavior from scores on the other related behavior. 9. A statistical intercorrelations among three or more behaviors, represented a. Partial Correlation by R. a. Partial Correlation b. Multiple Correlation b. Multiple Correlation c. Coefficient Correlation c. Both are correct. d. Multiple Regression Analysis d. None of the above. 15. A method in which different groups of subjects who are at different 10. It is an important correlation-based research method that can be used stages are measured at a single point in time; a method that looks for when subjects are measured on several related behaviors. time-related changes. a. Factor Analysis a. Cross-Lagged Panel Design b. Causal Analysis b. Cross-sectional Studies c. Path Analysis c. Ex Post Facto Studies d. None of the above. d. None of these. 11. A design in which the researcher compares the effects of different 16. It is also called direct relationship. a. Positive Correlation treatment conditions on preexisting groups of participants. b. Negative Correlation a. Quasi-Experimental Designs c. Both a and b. b. Cross-Lagged Panel Design d. None of the above. c. Nonequivalent Groups Design d. Longitudinal Design 17. A research design used to assess whether the occurrence of an events alters behavior; scores from measurements made before and after the 12. A study in which a researcher systematically examines the effects of pre- event are compared. existing subject characteristics (often called subject variables) by forming a. Crossed-Lagged Panel Design groups based on these naturally occurring differences between subjects. b. Nonequivalent Groups Design a. Cross-sectional Studies c. Longitudinal Design b. Correlation Study d. Pretest/Posttest c. Ex Post Facto Studies d. None of these. 18. A method in which the same group of subjects followed and measured at different points in time; a method that looks for changes across time.  a. Longitudinal Design 5. It is one inherent drawback of all correlational design, the problem of the b. Pretest/Posttest direction of cause and effect. c. Crossed-Lagged Panel Design 6. ____________ is a statistical intercorrelations among three or more d. Quasi-Experimental Designs behaviors, represented by R. 7. This analysis allows the statistical influence of one measured variable to 19. It can be used to explore the effects of different treatments on be held constant while computing the correlation between the other two. preexisting groups of subjects or to investigate the same kinds of 8. _____________ designs can seem like a real experiment, but they lack naturally occurring events, characteristics, and behaviors that we one or more of its essential elements. measure in correlational studies. 9. ____________is a formula for a straight line that best describes the a. Quasi-Experimental Designs relationship between the two variables. b. Longitudinal Design 10. A study in which a researcher systematically examines the effects of pre- c. Pretest/Posttest existing subject characteristics by forming groups based on these naturally d. None of the above. occurring differences between subjects. 11. A design in which the researcher compares the effects of different 20. It is Also called the line of best fit. treatment conditions on preexisting groups of participants. a. Linear Regression Analysis 12. _________ are used to establish relationships among preexisting behaviors and can be used to predict one set of behaviors from others. b. Regression Line 13. _________ is one that is designed to determine the correlation, or degree c. Both are correct. d. None of these. of relationship, between two traits, behavior, or events. 14. It represents the equation that best describes the mathematical relationship between two variables measured in a correlational study. 15. ___________ is a common correlation procedure that is used when individuals are measured on a large number of items. 16. It is a degree of relationship, between two traits, behavior, or events, represented by r. 17. ____________ is a correlation-based technique that uses regression equation to predict the score on one behavior from scores on the other related behavior. PRACTICE SET 2 (IDENTIFICATION) 18. _________ estimates the amount of variability in scores on one variable that can be explained by the other variable. 1. A research design used to assess whether the occurrence of an events alters behavior. 19. When two behaviors are strongly related, the researcher can estimate a 2. A method in which the same group of subjects followed and measured at score on one of the measured behaviors from a score from other. different points in time. 20. __________ an important correlation-based research method that can be 3. A method in which different groups of subjects who are at different used when subjects are measured on several related behaviors. stages are measured at a single point in time. 21. __________ is a relationship between two measures such that an 4. This design uses relationships measured over time to suggest the casual increase in the value of one associated with an increase in the value of path. the other.  22. __________is a relationship existing between two variables such that an 8. a increase in one is associated with a decrease in the other. 9. b 23. __________ are computed when the data from both measurements are 10. c ordinal, nominal, or one of the two is nominal. 11. c 24. The ____________ is used when interval or ratio scale data are collected. 12. c 25. The ___________is a visual representation of the scores belonging to 13. b each subject in the study. 14. d 15. b 16. a 17. d 18. a 19. a 20. b ANSWER KEYS PRACTICE SET 1 (MCQ) PRACTICE SET 2 (IDENTIFICATION) 1. a 2. b 1. Pretest/Posttest 3. a 2. Longitudinal Design 4. b 3. Cross-sectional Studies 5. b 4. Cross-Lagged Panel Design 6. d 5. Causal Analysis 7. b 6. Multiple Correlation 7. Partial Correlation 8. Quasi-experimental Designs 9. Regression Equation 10. Ex Post Facto Studies 11. Nonequivalent Groups design 12. Correlational Designs 13. Correlational Study 14. Regression Line 15. Factor Analysis 16. Correlation 17. Multiple Regression Analysis 18. Coefficient of Determination 19. Linear Regression Analysis 20. Path Analysis 21. Positive Correlation 22. Negative Correlation 23. Correlation Coefficient 24. Pearson r 25. Scatterplot