Since the first publication of psychometric data in 1998 (Measelle, Ablow, Cowan, & Cowan, 1998, Child Development), the Berkeley Puppet interview (BPI) has been included as a measure in over 100 published peer-reviewed studies, has been translated into 7 different languages, and has been identified as an “evidence based” assessment tool for preschoolers and children ages 4 to 8 years.
Overview of the BPI
Thank you for your interest in the Berkeley Puppet Interview (BPI). The BPI, building on a rich tradition of using puppets in clinical and research applications, was developed (Ablow & Measelle, 1993) to address the absence of standardized methodologies appropriate for measuring young children’s perceptions of themselves and their environments. Using an interactive technique for interviewing children, the BPI blends structured and clinical interviewing methods. During the actual BPI interview, two identical hand puppets (tan-colored puppy dogs named “Iggy” and “Ziggy”) make opposing statements about themselves and then ask children to describe themselves. For example,
Iggy: I have lots of friends.
Ziggy: I don’t have lots of friends
Iggy: How about you?
Ziggy: My parents’ fights are about me
Iggy: My parents’ fights are not about me
Ziggy: How about your parents?
Rather than use a forced-choice or recognition-task response format, the BPI allows children to respond in ways that are most natural and comfortable to them, with the goal of promoting a fluid and unselfconscious dialogue between a child and the puppets.
The majority of young children interviewed with the BPI respond verbally, either by describing themselves or by indicating which puppet is most like them. Other children use limited verbal responses, such as naming one of the two puppets, or by responding in non-verbal ways, such as pointing to a puppet. To capture the range of individual differences in young children’s responses, the BPI uses an extensive, rule-based coding system. Regardless of whether children’s responses are verbal, non-verbal, elaborated, or limited, the BPI’s coding system provides coders with the parameters needed to make sense of the varied ways that 4- to 8-year-old children respond to interview items. Specifically, guidelines have been developed to help coders decipher figures of speech, reasoning processes, and conditional responses that reflect ambivalent self-perceptions or uncertainty due to a lack of experience with a particular issue.
Published results demonstrate that the BPI is a reliable and valid measure of children’s perceptions. Children understand the questions and become unselfconsciously engaged in dialogue with the puppets, giving differentiated and coherent responses in the process. One of the key findings in these studies has been that the agreement between young children and adult informants tends to be as strong if not stronger than the level of agreement between pairs of adult informants. For example, agreement between children’s reports of their depressed feelings and teachers’ ratings of children’s internalizing behavior exceeded the level of agreement between teachers and mothers at 3 points in time (preschool, kindergarten, and first grade). Similarly, clinical observers’ ratings of marital conflict between parents were more highly correlated with 5- to 6-year-olds’ perceptions of their parents’ conflict than with parents’ own reports of their marital conflict. These data are important in light of the field’s tendency to (1) view young children’s perceptions as less valid and (2) rely on adult informants when attempting to understand young children’s subjective experiences or to identify emotional and behavioral problems in children younger than eight years of age.
At present, the BPI consists of several different interview protocols, each with multiple sub-scales. Thus far, the BPI’s utility has been tested on socioeconomically, culturally, and clinically diverse samples. Through collaborations with research laboratories in the United States and abroad, efforts to replicate and extend the BPI are underway. For example, a Spanish language version of the BPI Family Scales is being tested on a sample of low-income Chilean children and families.
BPI Domains & Scales
The BPI currently consists of a number of separate domains and scales: (1) the BPI Family Environment Scales assess children’s appraisals of central family relationships (parent-child, marital, sibling) and family processes (e.g., marital conflict, self-blame for conflict, shared-nonshared environment, perceived parental rejection); (2) the BPI Academic (BPI-A) and Social scales (BPI-Soc) tap children’s perceptions of teacher, school, and peers; (3) the BPI Symptomatology Scales (BPI-S) assess children’s perceived symptomatology in clinically relevant domains. Finally, Jeff Measelle and Oliver John (UC Berkeley) have re-classified the BPI’s self-perception items (from across multiple scales) and have extracted a fairly robust representation of the Big 5 personality factors, should this be of interest.
BPI Training Procedures
We ask that all research labs or clinical treatment centers interested in using the BPI undergo training in the method and receive certification. BPI workshops are typically 2 days in length, with follow-up practice interviews required for certification. Typically, we conduct BPI workshops at a researcher’s home institution. However, we also conduct 2-3 workshops per year at the University of Oregon.
Although we do not charge labs to use the BPI, we do charge for training and certification. A cost sheet is attached for your consideration (click here for Training Procedures). Please note that at this point in time we do not train individuals to train others in the method. Therefore, if researchers are interested in training several of their staff members, it usually makes the most sense to schedule or attend a training workshop.
To train a workshop, we typically need 2-3 months of advance notice to schedule the 2-day BPI training workshop and to budget the follow-up time needed to assist labs to achieve certification (following workshops, trainees complete 5-7 pilot BPI interviews, the videotapes of which are then sent to us for review and detailed feedback). We recommend that researchers not schedule or attend a BPI training workshop until (1) they have identified their staff and (2) will be ready to commence data collection once the certification process is complete. The whole training and certification process is much quicker if you have children ready to interview immediately following the workshop.
Follow-up Questions
Should you have follow-up questions or if you would like to receive copies of papers that have used the BPI, please email Drs. Ablow (jcablow@uoregon.edu) or Measelle (measelle@uoregon.edu). Again, thank you for your interest in the BPI.