Читать книгу Introduction to Abnormal Child and Adolescent Psychology - Robert Weis - Страница 120

Reproducibility is the scientific principle that the results of research studies must be replicated before they are accepted. The need for reproducibility is based on the notion that if a result is real, then another researcher should be able to obtain it using similar methods. Reproducibility is important because children and families are entitled to information that is trustworthy and accurate (Simons, 2015).

There are two types of replication studies: direct and conceptual. Direct replication involves repeating the exact procedures used in a research study to determine if the same results are found. Direct replication allows us to have confidence that the findings of the original study are accurate and robust. Conceptual replication involves using different methods to test the same hypothesis examined in the original study. Conceptual replication helps us confirm whether the theoretical ideas behind the original findings are true (Tackett, Brandes, King, & Markon, 2019).

Table 3.1

Note: Children who receive the measles, mumps, and rubella (MMR) vaccine get shots on at least 2 days between 6 and 24 months of age. Therefore, just by chance, we would expect 92 children in the United States each year to show the first signs of autism immediately after vaccination. Based on Howard and Reiss (2019).

For example, researchers might find that a phonics program improves the reading skills of children with dyslexia. The researchers might perform a direct replication by repeating the original study with a new sample of children with dyslexia using the same methods. Alternatively, they might perform a conceptual replication by using a different phonics-based reading program to see if it is also effective in helping children with dyslexia learn to read.

The Drug Abuse Resistance Education (D.A.R.E.) program illustrates the importance of reproducibility. D.A.R.E. was the most widely used alcohol and drug prevention program in the United States. Created in the 1980s, D.A.R.E. was designed to provide children with information about the dangers of drug use, teach decision-making skills, and build self-esteem. Weekly lessons were taught by police offers to students at school.

Early evaluations of D.A.R.E., conducted by researchers associated with the program, indicated that it was extremely successful (Faine & Bohlander, 1989). These results fueled the popularity of the program, which was eventually adopted by 80% of school districts and cost almost $1 billion annually (Ennett, Tobler, Ringwalt, & Flewelling, 1994).

Unfortunately, later studies were unable to reproduce the initial results. Overall, the students who received D.A.R.E. did not differ in their attitudes toward drugs, involvement with the police, or self-esteem from students who never participated in the program (West & O’Neal, 2004). More concerning, some children who participated in D.A.R.E. were more likely to use alcohol and other drugs during adolescence than children who did not participate in the program (Sloboda et al., 2009).

Although the early benefits of D.A.R.E. could not be replicated, the program continues to be offered in many school districts across the country. The program has been revised and rebranded. Some studies suggest that this new program may be more effective than the original D.A.R.E., but replication is needed before we can be confident in these findings (Caputi & McLellan, 2018).

The field of psychology devotes too little time and effort to reproducibility. Psychology journals tend to favor new and exciting research findings over replications. As a result, only about 1% of published research studies are replications (Makel, Plucker, & Hegarty, 2012).

Some psychologists have tried to make up for this shortcoming in replication research by attempting to reproduce the results of psychological studies published in well-respected journals. The largest attempt was led by Brian Nosek and the Open Science Collaboration (2015), a team of 270 researchers who performed direct replications of 100 studies in the fields of cognitive and social psychology. Unfortunately, the researchers were able to replicate only 36% of these studies. This low replication rate prompted some experts to claim that psychology is experiencing a replication crisis, that is, a failure to be able to reproduce its findings. Low replication rates have also been seen in other fields such as business, economics, and medicine (Martin & Clarke, 2018).

Recently, clinical psychologists have tried to encourage more replication research (Cybulski, Mayo-Wilson, & Grant, 2017). The Center for Open Science has created a repository where researchers can share data and report their replication studies. Similarly, the journal Perspectives on Psychological Science maintains a record of replication studies. Clinical researchers are also encouraged to preregister their treatment studies. Preregistration involves reporting the study’s purpose, design, and methods prior to observing its results. If the results show a treatment to be ineffective, then other researchers would have access to that important information (Nosek & Lindsay, 2018).

Review

 Pseudoscience consists of statements, beliefs, and practices that seem to be based on empirical evidence but are actually incompatible with scientific thinking. It capitalizes on our inherent cognitive and emotional biases.

 Science is a set of principles and procedures used to understand the natural world. It relies on careful, systematic collection of empirical data rather than expectations, emotions, or casual observations.

 Science is characterized by (1) falsifiability, (2) critical thinking, (3) parsimony, (4) precision, and (5) reproducibility.