Social Development Chapter 2 Research Methods: Tools for Discovery From theory to empirical support… ◼ Scientific Method Formulate a hypothesis based on theory ❑ Use replicable techniques to collect, study, and analyze data in an effort to test the theory’s usefulness ❑ ◼ In order to do this, researchers must Choose a research method ❑ Pick a research design ❑ Find a research sample ❑ Devise a data-collection strategy to test hypotheses in an effective and ethical way ❑ Formulating Hypotheses, Asking Questions ◼ Research starts with ideas derived from… Theory ❑ Previous research ❑ Observations ❑ A research study is only as good as the ideas behind it Formulating Hypotheses, Asking Questions ◼ Steps between ideas and conducting a good study… ❑ Translate the general idea into clear research hypotheses or questions ❑ Find out about past work on the topic ❑ Operationalization – Defining a construct so that it is observable and measurable ◼ ❑ Construct – An idea or concept, especially a complex one such as aggression or love Make decisions about research designs, methods, samples, and analyses Establishing Patterns and Causes ◼ The Correlational Method ❑ Involves determining whether two variables are related to each other in a systematic way and how strongly they are related ❑ A correlation between two variables does not indicate anything about cause and effect Establishing Patterns and Causes ◼ Correlation Coefficient Numerical estimate of how closely two variables are related to each other and the direction of the relation ❑ Ranges from +1.0 to -1.0 ❑ ◼ The closer to 1.0, the stronger the relation; the closer to 0, the weaker the relation A -1.0 correlation means that for every increase in one variable there is a systematic decrease in the other ❑ A +1.0 correlation means that for every increase in one variable, there is a systematic increase in the other +1.00 0 ❑ -1.00 Why use correlational research? ◼ We cannot always design a suitable experiment to study our question ❑ Some variables of interest cannot be studied in a controlled way ◼ Ethical concerns sometimes prohibit using experiments ❑ e.g., cannot randomly assign families to abusive and non-abusive groups ◼ Understanding causal processes is not the only goal of research Establishing Patterns and Causes ◼ Laboratory Experiments Primary way to study causal connections ❑ Researchers hold constant, or equate, every possible factor except the one they have hypothesized will influence the behavior they want to study ❑ Researchers randomly assign each participant to a group ❑ Establishing Patterns and Causes ◼ Experimental group – participants exposed to the proposed causal factor ◼ Control group – participants not exposed to the proposed causal factor ◼ Random assignment – rules out the possibility that the people in the groups differ from one another in some systematic way that could distort the results of the experiment Establishing Patterns and Causes ◼ Independent Variable (IV) – the factor that researchers deliberately manipulate in an experiment ◼ Dependent Variable (DV) – the factor that researchers expect to change as a function of change in the independent variable Establishing Patterns and Causes Laboratory experiments may lack ecological validity, that is, they may not accurately represents events or processes that occur in the real world ◼ An alternative is the laboratory analogue experiment ◼ ❑ Attempt to duplicate in the laboratory features or events that occur naturally in everyday life in order to increase the ecological validity of the results Research strategies must be matched to the questions researchers are asking Establishing Patterns and Causes Field Experiment – experiment in which researchers deliberately create a change in a real-world setting and then measure the outcome of their manipulation ◼ Intervention – program provided to improve a situation or relieve psychological illness or distress ◼ Natural Experiment (Quasi Experiment) – experiment in which researchers measure the results of events that occur naturally in the real world (e.g., effects of a hurricane, adoption, residence change) ◼ Combining Different Methods ◼ All research designs have their strengths and weaknesses Correlational designs may be a good starting point in an area not well known; then follow up with an experiment to test causal relations ❑ May be best to combine laboratory and field approaches in a single study to improve generalizability of findings ❑ Combining Different Methods ◼ ◼ ◼ Traditional laboratory experiment (cell A) Typical field experiment (cell D) IV introduced in the laboratory and the DV measured in the field (cell C) IV introduced in the field and the DV measured in the lab (cell B) Manipulation of IV Assessment of DV ◼ Lab Field Lab A B Field C D Case Study Approach ◼ A form of research in which investigators study an individual person or group intensely Allows exploration of rare phenomena ❑ Facilitates intensive investigation ❑ Provides rich details ❑ May serve as a precursor to follow-up studies using alternative designs and methods ❑ Limitation: Impossibility of generalization Studying Change Over Time ◼ Cross-sectional design – design in which researchers compare groups of individuals of different age levels at approximately the same point in time ◼ Disadvantages: Design yields no information about the causes behind age-related differences ❑ No information about stability or change of individuals across age ❑ ◼ Could reflect age-related change or historical change Studying Change Over Time ◼ Longitudinal Design – study in which investigators follow the same people over a period of time, observing them repeatedly Allows for examination of changes in individuals across time rather than simply differences across age groups ❑ Allows for examination of stability over time ❑ Allows for examination of possible causes of change over time ❑ Longitudinal Design: Examination of Causes of Change ◼ Growth curves of five children in a longitudinal study ❑ ◼ When the children were removed from an abusive home environment (*), they immediately gained in height. This study provides clear evidence that abuse inhibits growth. Sirotnak, 2008. Image reprinted with permission from eMedicine.com, 2010. Available at: http://emedicine.medscape.com/article/913843overview Longitudinal Design: Disadvantages ◼ Not as efficient as cross-sectional approach ◼ Loss of participants over the course of the study ❑ Problem of representativeness ◼ Lack of flexibility in terms of updating methods of measurement ◼ Practice effects ◼ Cohort effects ◼ More time consuming and expensive Studying Change Over Time ◼ Sequential Design A way of studying change over time that combines features of both cross-sectional and longitudinal designs Sequential Design: Advantages ◼ Can examine age-related changes in children’s behavior and look at the stability of individual differences ◼ Can examine practice effects because although some children are assessed multiple times, other children are assessed less often ◼ Can examine cohort or generational effects Comparison of Methods of Studying Developmental Change over Time Cross-Sectional Longitudinal Sequential Time required for study is relatively Short Long Moderate Cost of study is likely to be Low High Moderate Participant attrition during study is most likely Low High Moderate Risk of staff turnover is Low High Moderate Possibility of using newest measures is High Low Moderate Likelihood of practice effect is None High Low or Moderate Ability to assess links between early events and later behavior is None High Ability to assess stability vs. instability of behavior is None High Ability to assess developmental paths of individuals is None High Unknown High High for ShortTerm High for ShortTerm High for ShortTerm Moderate Threat of a cohort effect tends to be Selecting a Sample ◼ Representative Sample A research sample in which participants are drawn from strata or categories (e.g., social classes or ethnic groups) in the same proportions as they are found in the larger population ❑ Avoids problems associated with biased or skewed samples ❑ Allows for generalization ❑ Selecting a Sample ◼ The National Survey Approach Researchers select a large, nationally representative group of participants ❑ Use of oversampling for specific groups ❑ Allows researchers to draw conclusions that apply to entire populations and subpopulations ❑ But… ❑ ◼ ◼ ◼ Costly in terms of time and labor Surveys are not well suited to answering questions about the psychological processes underlying social development Using subsamples from the national sample to do more in-depth research can be useful Selecting a Sample ◼ Meta-analysis ❑ A statistical technique that allows the researcher to summarize the results of many studies on a particular topic and to draw conclusions about the size and replicability of observed differences or associations ❑ Yields an overall estimate, or effect size, of the magnitude of the difference between experimental and control groups or the strength of associations between factors in correlational studies Note that the accuracy of that estimate is limited by the samples in the individual studies and the quality of the designs Studying Development Cross-Culturally ◼ Selecting samples from different cultures can be a valuable research strategy Behavior pattern similar across cultures → universal ❑ behavior patterns differ across cultures → specific environmental variables play a significant role ❑ ◼ Difficult and expensive to conduct ◼ Potential problems of researcher bias Gathering Data ◼ Children’s self-reports – Information child provides about herself by answering the researchers’ questions ❑ Special problems ◼ ❑ Children less attentive, slower to respond, and have more trouble understanding the questions that researchers ask Special methods ◼ Puppet interview technique ◼ Story completion ◼ Narratives ◼ Experience sampling method – participants are signaled at random times throughout the day and record answers to researchers’ questions; also called the beeper method Courtesy of Jennifer Ablow and Jeffrey Measelle, University of Oregon A sample answer sheet completed by a child using the experience sampling method. The child completed these questions when “beeped” throughout the day (Source: Vandell et al., 2005.) Vandell, D. L., Shernoff, D. J., Pierce, K. M., Bolt, D. M., Dadisman, K., & Brown, B. B. (2005). Activities, engagement, and emotion in after-school programs and elsewhere. New Directions for Youth Development, 105, 121–129. This material is reproduced with permission of John Wiley & Sons, Inc. Gathering Data ◼ Reports by family members ❑ ❑ ❑ Parent reports are generally based on many observations made over time in a variety of situations Even if not totally accurate, parents’ perceptions, expectations, beliefs, and interpretations of events and behavior may be just as important as objective reality Ways to increase accuracy ◼ ◼ ◼ ◼ Ask only about very recent events Ask parents to keep a structured diary Ask about “family stories” Reports by teachers and peers Gathering Data ◼ Focus Group – Group interview in which an interviewer poses questions that are answered by the participants Valuable in early stages of research ❑ Can increase the validity of a study ❑ Can help researchers construct questionnaires ❑ May also be useful at the end of the study as a means of making sense of findings ❑ Gathering Data ◼ Direct Observation – Researchers go into real settings or bring participants into the laboratory to observe behaviors of interest ◼ Naturalistic Observation – Observations collected in the child’s natural settings without interfering with the child’s activities ❑ ❑ ❑ Problems with infrequently occurring behavior Problems of reactivity – The change in a person’s behavior due to the fact that he or she is being observed Observer bias – An observer’s tendency to be influenced by knowledge about the research design or hypothesis Gathering Data ◼ Structured observation – Observation in which researchers create a situation so that behaviors they wish to study are more likely to occur (e.g., clean-up task to assess child compliance) ❑ Avoids problems with infrequently occurring behavior ❑ But, concern with ecological validity ◼ Can make the lab more “home like” Ways of Recording and Coding Observations ◼ Behavior observations ❑ ❑ ❑ Specimen record – researchers record everything a person does within a given period of time Event sampling – researchers record participants’ behavior only when an event of particular interest occurs Time sampling – researchers record any of a set of predetermined behaviors that occur within a specified period of time Ways of Recording and Coding Observations ◼ Ethnography – Use of intensive observations and interviews to gather data about the beliefs, practices, and behaviors in a particular context or culture ❑ Participant observations – research strategy used to gain familiarity with a group of individuals (e.g., religious, subcultural group) by means of intensive involvement in their activities, usually over an extended period of time Ways of Recording and Coding Observations ◼ Drawbacks of Ethnography ❑ May offer subjective views ❑ Practices of group may be disrupted by outside observer ❑ Researchers field notes may be subjective and biased ❑ Limited generalizability ◼ Best when used in combination with other research methods Ways of Recording and Coding Observations ◼ Nonverbal measures Particularly important in the study of babies and young children ❑ Examples ❑ ◼ ◼ ◼ ◼ Sucking response Visual preference Smiling, vocalization, crying Habituation – An individual reacts with less and less intensity to a repeatedly presented stimulus until he or she responds only faintly or not at all Ways of Recording and Coding Observations ◼ Internal Responses ❑ Measurement of psychophysiological responses Physiological bases of psychological processes measured by brain activity, brain waves, and heart rate Alexander Tsiaras/Science Source Table 2-2 Techniques for Studying Human Brain Function and Structure EEG (electroencephalography) What It Is Recording of the brain’s spontaneous electrical activity over a short period of time by means of multiple electrodes placed on the scalp. Advantage Detects very rapid changes in electrical activity allowing analysis of stages of cognitive processing Disadvantage Provides poor spatial resolution of the source of electrical activity PET (positron-emission tomography) and SPECT (single- photon emission tomography) What It Is A visual image of an injected radioactive substance showing blood flow or glucose use reflecting changes in neuronal activity Advantage Provides spatial resolution better than EEG but less than MRI Disadvantage Cannot follow rapid changes (faster than 30 seconds); requires exposure to low levels of radioactivity Table 2-2 (continued) MRI (magnetic resonance imaging) and fMRI (functional magnetic resonance imaging) What It Is MRI is used to provide high spatial resolution of brain anatomy; fMRI is used to provide images of changes in blood flow that indicate specific anatomical details and changes in neural activity Advantage Provides high temporal resolution; requires no exposure to radioactivity Disadvantage High cost to operate TMS (transcranial magnetic stimulation) What It Is Shows which brain regions are necessary for given tasks by changes after TMS is applied to a location Advantage Temporarily disrupts a specific region of brain by exposing it to intense magnetic energy Disadvantage Long-term safety not well established Ways of Recording and Coding Observations ◼ Internal Responses ❑ Measurement of hormone levels in the body ◼ ◼ Hormone – Powerful and highly specialized chemical substance produced by the cells of certain body organs, which has a regulatory effect on the activity of certain other organs Cortisol – A hormone secreted by the adrenal glands in response to physical or psychological stress Analyzing Data ◼ Qualitative study – uses nonstatistical analysis of materials to gain in-depth understanding ❑ Not trying to confirm or falsify a hypothesis ❑ Subjective summary of findings may be final step or may precede designing a follow-up quantitative study Analyzing Data ◼ Quantitative study – Research involving statistical analysis of numerical data Summarize numbers using descriptive statistics ❑ Determine whether findings are statistically significant or merely due to chance ❑ Analyses can range from simple statistics (e.g., ANOVA, t test) to more complex approaches that examine multiple variables simultaneously (e.g., SEM) ❑ Table 2-3 Children’s Rights in Social Development Research 1. The right to be fully informed. Every child has the right to full and truthful information about the purposes and procedures of a study in which he or she is to participate. 2. The right to give informed and voluntary consent. Every child has the right to agree, orally or in writing, to participate in a research project. If a child is too young to understand the study and make an informed decision, researchers must obtain the informed consent of the child’s parents or those who act in loco parentis such as teachers or camp counselors who are temporarily responsible for the child. 3. The right not to be harmed in any way. Every child has the right not to experience physical or psychological harm as a result of the research procedures. 4. The right to withdraw voluntarily from research. Every child has the right to withdraw from participation in the study at any time. Table 2-3 (continued) 5. The right to be informed of the results of research. Every child has the right to information about the results of the research. If the child is too young to fully understand, the information must be provided to the child’s parents. 6. The right to confidentiality. Every child has the right to know that personal information gathered as part of the research will remain private and confidential and will not be shared with any other individuals or agencies. 7. The right to full compensation. Every child has the right to be compensated for time and effort as a research participant even if he or she withdraws from the study. Incentives must be fair and not exceed the range children normally receive; this precaution ensures that incentives are not used to coerce the child to participate. 8. The right to beneficial treatments. Every child has the right to profit from any beneficial treatments provided to other participants in the research. When experimental treatments are deemed beneficial, for example, participants in control groups have the right to the same beneficial treatment after the project is completed. (Sources: Based on reports from the American Psychological Association [2002] and the Society for Research on Child Development, Committee on Ethical Conduct in Child Development Research [2007].) Concluding Thoughts ◼ The scientific method is used to answer questions or test theory-based hypotheses ◼ Developmental researchers need to make decisions about the general design they will use (correlational, experimental) as well as the way in which they will study change over time (cross-sectional, longitudinal, sequential) ◼ Decisions also need to be made about data collection and analysis methods ◼ All approaches have strengths and weaknesses and a combination of approaches is usually better than any single approach ◼ In any study, the ethical risks or costs of the study to the participants needs to be carefully weighed against the potential gains
Purchase answer to see full attachment