Bandura Evaluation

• GENERALISING BEYOND THE SAMPLE

  '-' The findings may not be generalizable to participants from lower socioeconomic backgrounds. The children were all from the nursery of a prestigious university, which was for children of staff and students. Therefore, these children may have had particularly educated parents and their home lives may not be representative of the rest of the population. Their learning experiences may have differed from other children and they may have been more or less likely to imitate adult behavior as a result, reducing the validity of the findings.
• GENERALISING TO EVERYDAY LIFE

  '-' The situation presented to the children lacks mundane realism as it was an unusual set-up. An adult will rarely be seen to attack a toy, with the child being provided with the opportunity to imitate this behavior. The children were seen hitting an inflatable doll rather than a real person. Bobo dolls are also designed to be hit and bounce back. Therefore, it is not possible to assume that the aggressive behavior shown toward the toys would extend to violence toward real people. Bandura et al.'s study did not distinguish between play fighting and violence in real life.
• STANDARDISATION


  '+' The standardised procedure and instructions, which increases the reliability of the study. The layout of the toys in the experimental room was kept the same for participants in the aggressive, non-aggressive and control conditions. The length of time the children were observed for was also the same, at 20 minutes, and the same behavioural checklist was used to record behaviour every live seconds. This level of consistency makes the study's findings about children's delayed

  imitation of aggression more reliable.
• INTER-OBSERVER RELIABILITY


  '+' High levels of inter-observer reliability were found in one-half of the trials, two observers independently recorded participants' behavior using a behavioral checklist in five-second intervals. Their results were correlated and found to always be in the +0.90s. This ensured that there was a high level of consistency in the scoring of aggressive and non-aggressive behaviors. There was strong inter-rater reliability between the researcher and the nursery teacher for the children's aggression scores. The raters' scores correlated +0.89.

  This indicates a high level of consistency in how the children were rated across the aggression scales.
• APPLICATIONS TO EVERYDAY LIFE: TELEVISION WARNINGS

  As Bandura's study demonstrated both physical and verbally aggressive behavior can be observed and imitated by children, it is important for television networks to either consider censoring content or provide warnings when there will be aggressive content on the screen so that parents have the option to prevent their children from viewing it. The producers of children's television shows can ensure that characters are either non-aggressive or that aggressive content is reduced to a minimum.
• APPLICATIONS TO EVERYDAY LIFE: PARENTS

  Bandura demonstrated that children were more likely to imitate a male model, so parents may wish to proactively expose their children to friendly and prosocial male role models. Additionally, children were less likely to show aggressive behaviors if they observed a non-aggressive model, so parents can direct their children to watch television programmes with models who demonstrate prosocial behaviors.
• MATCHED AGGRESSION SCORES

  '+' Participants were matched on prior aggression levels.

  They were each rated by their nursery teacher and the experimenter independently on four five-point rating scales for their: physical aggression, verbal aggression, aggression towards inanimate objects, and aggressive inhibition. Using the total scores gained, participants were put into triplets, with each being randomly assigned to a different condition. This increased the validity of the findings as individual differences in prior aggression levels should not have caused any differences between conditions.
• ONLY TWO STOOGES


  '-' Only one stooge was used in each male and female condition throughout the experiment. This means that the children may have imitated the model due to some individual feature that was unique to the model, rather than their sex.

  This confounding variable may have affected how frequently a model was imitated and thus the validity of the results.
• ETHICAL ISSUES


  '-' The risk to the children of psychological harm. One-third of the participants witnessed the aggressive behavior of a model. This group of children was subsequently more likely to imitate this aggressive behavior. Since this reproduction was after a delay, it is possible that the children continued to demonstrate delayed imitation of the aggressive behavior even after the study had ended or to feel in an aggressive state of mind when they returned home.
• QUANTITATIVE DATA


  '+' Quantitative data was collected in this study, which allows for more objective comparisons to be made between conditions of the independent variables. Data was collected on the number of aggressive acts shown by each child using a

  behavioral checklist. Therefore, it was possible to compare numerical data regarding the children's levels of aggression following exposure to different types of models without bias, increasing validity.
• INSIDE INFORMATION


  '-' The main observer for all of the trials was the male model from the experiments. This meant that he knew which children had been in the male model conditions and may have been more subjective in his interpretation of their behaviors. This would subsequently reduce the validity of the data for the male conditions as he may have expected the children he had previously seen to act in a certain way depending on whether they had watched him acting aggressively and interpreted their behavior accordingly.
• SINGLE-BLIND


  '+' The main observer did not know which condition each child had been in for any trials for which he had not been the model

  which would increase his objectivity. This means that any similarities or differences between the female conditions were

  not due to bias on the part of the researcher which increases the validity of the data for these conditions.
• THE USE OF CHILDREN IN PSYCHOLOGICAL

  RESEARCH

  This study used young children aged 37 to 69 months, which might make them particularly vulnerable to demand characteristics. The young children may believe that the adults expect them to copy the behavior of the stooge and so act accordingly. This would cause issues with the validity of the results. The children are also at a sensitive age in their development, which raises ethical issues for those being exposed to an aggressive model, as there may be a greater risk of psychological harm due to long-term after-effects.
• GENERALISING BEYOND THE SAMPLE

  '-' The findings may not be generalizable to participants from lower socioeconomic backgrounds. The children were all from the nursery of a prestigious university, which was for children of staff and students. Therefore, these children may have had particularly educated parents and their home lives may not be representative of the rest of the population. Their learning experiences may have differed from other children and they may have been more or less likely to imitate adult behavior as a result, reducing the validity of the findings.
• GENERALISING TO EVERYDAY LIFE

  '-' The situation presented to the children lacks mundane realism as it was an unusual set-up. An adult will rarely be seen to attack a toy, with the child being provided with the opportunity to imitate this behavior. The children were seen hitting an inflatable doll rather than a real person. Bobo dolls are also designed to be hit and bounce back. Therefore, it is not possible to assume that the aggressive behavior shown toward the toys would extend to violence toward real people. Bandura et al.'s study did not distinguish between play fighting and violence in real life.
• STANDARDISATION


  '+' The standardised procedure and instructions, which increases the reliability of the study. The layout of the toys in the experimental room was kept the same for participants in the aggressive, non-aggressive and control conditions. The length of time the children were observed for was also the same, at 20 minutes, and the same behavioural checklist was used to record behaviour every live seconds. This level of consistency makes the study's findings about children's delayed

  imitation of aggression more reliable.
• INTER-OBSERVER RELIABILITY


  '+' High levels of inter-observer reliability were found in one-half of the trials, two observers independently recorded participants' behavior using a behavioral checklist in five-second intervals. Their results were correlated and found to always be in the +0.90s. This ensured that there was a high level of consistency in the scoring of aggressive and non-aggressive behaviors. There was strong inter-rater reliability between the researcher and the nursery teacher for the children's aggression scores. The raters' scores correlated +0.89.

  This indicates a high level of consistency in how the children were rated across the aggression scales.
• APPLICATIONS TO EVERYDAY LIFE: TELEVISION WARNINGS

  As Bandura's study demonstrated both physical and verbally aggressive behavior can be observed and imitated by children, it is important for television networks to either consider censoring content or provide warnings when there will be aggressive content on the screen so that parents have the option to prevent their children from viewing it. The producers of children's television shows can ensure that characters are either non-aggressive or that aggressive content is reduced to a minimum.
• APPLICATIONS TO EVERYDAY LIFE: PARENTS

  Bandura demonstrated that children were more likely to imitate a male model, so parents may wish to proactively expose their children to friendly and prosocial male role models. Additionally, children were less likely to show aggressive behaviors if they observed a non-aggressive model, so parents can direct their children to watch television programmes with models who demonstrate prosocial behaviors.
• MATCHED AGGRESSION SCORES

  '+' Participants were matched on prior aggression levels.

  They were each rated by their nursery teacher and the experimenter independently on four five-point rating scales for their: physical aggression, verbal aggression, aggression towards inanimate objects, and aggressive inhibition. Using the total scores gained, participants were put into triplets, with each being randomly assigned to a different condition. This increased the validity of the findings as individual differences in prior aggression levels should not have caused any differences between conditions.
• ONLY TWO STOOGES


  '-' Only one stooge was used in each male and female condition throughout the experiment. This means that the children may have imitated the model due to some individual feature that was unique to the model, rather than their sex.

  This confounding variable may have affected how frequently a model was imitated and thus the validity of the results.
• ETHICAL ISSUES


  '-' The risk to the children of psychological harm. One-third of the participants witnessed the aggressive behavior of a model. This group of children was subsequently more likely to imitate this aggressive behavior. Since this reproduction was after a delay, it is possible that the children continued to demonstrate delayed imitation of the aggressive behavior even after the study had ended or to feel in an aggressive state of mind when they returned home.
• QUANTITATIVE DATA


  '+' Quantitative data was collected in this study, which allows for more objective comparisons to be made between conditions of the independent variables. Data was collected on the number of aggressive acts shown by each child using a

  behavioral checklist. Therefore, it was possible to compare numerical data regarding the children's levels of aggression following exposure to different types of models without bias, increasing validity.
• INSIDE INFORMATION


  '-' The main observer for all of the trials was the male model from the experiments. This meant that he knew which children had been in the male model conditions and may have been more subjective in his interpretation of their behaviors. This would subsequently reduce the validity of the data for the male conditions as he may have expected the children he had previously seen to act in a certain way depending on whether they had watched him acting aggressively and interpreted their behavior accordingly.
• SINGLE-BLIND


  '+' The main observer did not know which condition each child had been in for any trials for which he had not been the model

  which would increase his objectivity. This means that any similarities or differences between the female conditions were

  not due to bias on the part of the researcher which increases the validity of the data for these conditions.
• THE USE OF CHILDREN IN PSYCHOLOGICAL

  RESEARCH

  This study used young children aged 37 to 69 months, which might make them particularly vulnerable to demand characteristics. The young children may believe that the adults expect them to copy the behavior of the stooge and so act accordingly. This would cause issues with the validity of the results. The children are also at a sensitive age in their development, which raises ethical issues for those being exposed to an aggressive model, as there may be a greater risk of psychological harm due to long-term after-effects.
• What is a strength of the research method used?

  Lab experiment- high control over extraneous variables high internal validity- the cause and effect between role model and aggression levels can be established
• What is a weakness of the research method used?

  Low in ecological validity- the task and setting are not representative of real life so limits the generalizability of the study to real life
• What is a strength of the sample used

  - sample is not andocentric or gynocentric as there is an equal number of boys and girls
• What is a weakness of the sample used

  - use of only children from one culture who attend Stanford university nursery and so not representative of children who go to other nursery universities

  - the experimenter could have picked aggressive children specifically
• What is a strength of the type of data used

  - quantitative data- can easily be analyzed and make much easier judgments in which condition caused more aggression and provides greater insight into the children's gender specific ideas
• What is a weakness of the type of data used

  -qualitative data- can't make direct judgments on the comments made "that's not the way for a women to behave" "that man is a strong fighter"
• How is the validity high

  High internal validity due to the controlled lab experiment so can control or extraneous variables
• How is e validity low

  Low internal validity - as the female experimenter could have missed signs of aggressive behaviour could have missed signs of aggressive behaviour when observing behaviour in the playground
• How is the reliability high

  High internal reliability- as standardized procedures were used- in room 1 the procedure was replicated 3 times for each child
• How is the study reductionist

  Bandura only investigates aggressive behaviour only being caused by imitations and ignoring other factors like the children's personality
• How is the study holistic

  - looks at multiple aspects of aggression considered such as physical, verbal
• How does the study link to nurture

  Children learn/ acquire aggressive behaviour if they are surrounded by it from a role model like parents
• How does the study like to mature

  Biological stages 4/5 are more likely to be influenced and not developed morals
• How is the study useful

  The study enhances our knowledge of how children image aggressive behaviour from their role models and practical applications can be made as to how behave in a positive manner