Chp 1B Research Methods.pdf

Chapter 1B: Research in Child Development Doing Child-Development Research Child-Development Research and Family Policy Children and Their Development, Cdn Ed. - Robert Kail  Doing Child-Development Research Measurement in Child-Development Research General Designs f...

Chapter 1B: Research in Child Development Doing Child-Development Research Child-Development Research and Family Policy Children and Their Development, Cdn Ed. - Robert Kail  Doing Child-Development Research Measurement in Child-Development Research General Designs for Research Designs for Studying Age-Related Changes Ethical Responsibilities Communicating Research Results © 2022 Pearson Education Canada Inc.  Identifying the Problem Scientific method: rules for designing, conducting, evaluating, and communicating research Hypotheses vs. Theories – An hypothesis is a statement of what you think should happen in your experiment – A theory is a set of integrated statements that organizes/predicts results and is more encompassing than an hypothesis – Operational definition (variables) Baird Measurement in Child-Development Research Measures should be both valid and reliable Samples of children who participate in research should be representative of the population of interest  Developmental Research Main issues Research Methods & Design Descriptive (Observation, Self-report ) Correlation Experiment Ethical procedures Measurement in Child-Development Research Two types of Systematic Observation: – Naturalistic observation observe children in their natural environment – Structured observation Create/stage a setting to elicit behaviours of interest Sampling behavior with tasks – When direct observation is difficult Self reports – Children are asked to reflect on a topic – Include questionnaires and interviews  Descriptive Research Interview methods – Open-ended vs. structured – Can provide a wealth of information – Rely on the informant’s knowledge, memory and ability or willingness to communicate Case studies – Bring a wide range of information on one child – Often are concerned with clinical issues – Cautions need to be made when drawing conclusions  Descriptive Research Correlational Studies – Measures the relation between variables as they occur naturally in the world Provides an index termed the correlation coefficient (r) Indicates the strength (0-1) and direction (+, -) of the relation Correlational Research Correlational Research: Example A high positive r (0.78) does not imply that watching Sesame Street causes improved reading – only that the two variables are related Suggests the need for an experiment Three Interpretations of a Correlation Coefficient © 2022 Pearson Education Canada Inc.  Experimental Research Experiments offer the opportunity for causality, i.e., manipulation of one variable induces change in another variable Variables in an experiment: – Independent variable (IV) is manipulated by the experimenter and is assumed to be a causal factor – Dependent variable (DV) is measured by the experimenter and is assumed to be controlled by the IV Example of an Experiment An experiment tests a null hypothesis and uses inferential statistics to draw conclusions © 2022 Pearson Education Canada Inc.  Experimental Research: Example of a Group Study Hypothesis: IV: DV: Four groups are asked to watch either 2, 4, 8, or 10 hours per week Sesame Street Viewing and Reading Test Results: Increases in Performance. Each bar shows the average test score viewing time lead to for all the children in that experimental group. What increases in reading do these results indicate about the relation between Sesame Street viewing and reading ability? Why can level we say from these findings that viewing Sesame Street improves reading ability?  Experimental Research: Example of a Group Study Hypothesis: Watching Sesame Street improves reading ability IV: Hours of watching DV: Reading level test scores Four groups are asked to watch either 2, 4, 8, or 10 hours per week Results: Increases in viewing time lead to increases in reading level Figure 2.3 Sesame Street Viewing and Reading Test Performance. Each bar shows the average test score for all the children in that experimental group. What do these results indicate about the relation between Sesame Street viewing and reading ability? Why can we say from these findings that viewing Sesame Street improves reading ability?  Experimental Research: Other Experimental Designs Sometimes a group is not available for a study – Study can be conducted using one subject – Reversal Replication Design A B A B Sometimes treatment CANNOT be reverted or withdrawn  Reversal-Replication Design A reversal-replication design showing a causal relation between the presence of mother (the independent variable) and the amount an infant smiles (the dependent variable). The third and fourth phases replicate the procedures and results of the first two phases.  ABAB 1  Multiple Baselines 2  Experimental Research: Other Experimental Designs Sometimes variables cannot be manipulated – Ethical concerns – The nature of some IVs (cultural background) – Quasi-experimental studies allow researchers to compare groups differing on some important characteristics Special Designs for Studying Age- Related Changes Longitudinal study: the same individuals are tested repeatedly over time Cross-sectional study: children of different ages are tested Longitudinal-sequential studies are hybrids of longitudinal and cross-sectional studies Microgenetic study: children are tested repeatedly over a span of days or weeks © 2022 Pearson Education Canada Inc.  Special Designs Longitudinal Studies – Observes or tests group/s of individuals over a long period of time (stability estimates/impact of early experience) Cost, time, attrition, practice Cross-Sectional Studies – Observes or tests groups of various ages for differences More time-effective, less expensive, but cannot tap into stability of behaviour (different Ss are used) May involve a cohort effect  Longitudinal Study Figure 1.6 A longitudinal study—the same children are tested across a span of six years.  Cross-Sectional Study Figure 1.7 A cross-sectional study: Children at different ages are tested at the same time  Special Designs Longitudinal-Sequential Studies A combination of cross-sectional and longitudinal designs Allows for flexibility to collect information in several ways Avoids “cohort effects”  Longitudinal-Sequential Study Figure 1.8 A longitudinal-sequential study: children at different ages are tested across several years.  A Longitudinal-Sequential Design A cross-sequential design. Children aged 4, 7, and 10 years are compared cross-sectionally at Time 1 to reveal age differences. Three years later, at Time 2, another cross-sectional design is conducted comparing these children who are now aged 7, 10, and 13. How might cohort effects play a role in any differences found? To examine the stability of competitiveness, each of the three groups is examined longitudinally comparing their scores at Time 2 to those at Time 1. How could repeated testing affect the results? Comparing the groups who were 7 and 10 at Time 1 with those who are 7 and 10 at Time 2 allows the researcher to examine for a cohort effect or a repeated testing effect.  Special Designs Microgenetic studies examine developmental changes as they are occur; children are tested repeatedly over a span of days or weeks Studies small numbers of children and administers repeated measures at the time the developmental change is expected to occur – Costly – Repeated assessment itself may cause change  Other Research Methods Cross-cultural research examines a behaviour under different cultural conditions Comparative research examines a behaviour in non-humans Psychophysiological methods – Focus on a particular aspect of behaviour (e.g. attention or stress) by measuring heart rate or cortisol levels – Brain activity as children perform specific tasks  Ethical Responsibilities Guidelines for Ethical Research in Canada: Try to conduct research which benefits humanity Minimize risks to research participants Informed consent and Assent Avoid major deception Keep results anonymous or confidential Provide debriefing afterward Communicating Research Results Research results are reported in scientific journals Converging evidence from many studies is necessary © 2022 Pearson Education Canada Inc.  Child-Development Research and Family Policy Family Policy refers to any legal policy that directly affects families with children: Infant mortality/poverty/abuse/suicide/drug use Family policy must be informed by factual knowledge from research, not based on stereotypes or common myths Other Examples – Television/video game rating systems – Child Support © 2022 Pearson Education Canada Inc. Why Link Research to Family Policy? In the field of Applied developmental science research must be directly applicable to the promotion of healthy development Particular attention is given to ‘at risk’ children and families © 2022 Pearson Education Canada Inc.  Why Link Research to Family Policy? The concept of ‘family’ has changed over time We need to redefine policy to reflect these changes Research allows us to better understand the effects of these changes and how to propose appropriate policy change © 2022 Pearson Education Canada Inc.  Ways to Influence Family Policy Build understanding of child-development Researchers have an obligation to inform the public and policy-makers Family policy must be informed, not based on stereotypes or common myths © 2022 Pearson Education Canada Inc.  Ways to Influence Family Policy Serve as an advocate for children Children cannot represent their own interests Advocacy groups act to influence policy on specific issues  Ways to Influence Family Policy Evaluate policies and programs Child-development experts evaluate policies which indirectly affect children Example: – Changes to the welfare system Researchers have the tools to explore potential effects of these policies before they are put in place © 2022 Pearson Education Canada Inc.  Ways to Influence Family Policy Develop a model program Before considering a large-scale program, policy-makers like to see results at a smaller level Example: – A breakfast program at a local school may serve as a model for a provincial program © 2022 Pearson Education Canada Inc.  An Emphasis on Policy Implications Improves Research The relation between policy and research is mutually beneficial Researchers tend to focus on the scientific and academic aspects of their research Researchers benefit greatly from being forced to see the ‘the bigger picture’ and apply their research to ‘the real world’ © 2022 Pearson Education Canada Inc.