The key limitations in this research were self report bias and attrition bias caused by parental withdrawal. These limitations affected the reliability and validity of the research in different ways.
Self report bias
The research relied upon self report assessments to establish relationships between the parameters. Subjects filled in assessment forms during and after the intervention but biases affected the reliability of the research.
The main challenge with this method of data collection is it is difficult to control the level of honesty among the participants. Some of them may not be honest about certain subjects that they perceive to be personal.
This bias is difficult to accommodate in the research analysis because one cannot assess a participant’s honesty levels. It all depends on their personality and their interpretations of the research parameters. In addition to intentional dishonesty, some of the participants were liable to introspectability.
In self report research, subjects may perceive themselves in a totally different way than is the case. Even if that participant may genuinely want to be honest, he or she may not know that he or she is representing an incorrect view of himself or herself (Miller & Wright, 1995).
Certain individuals lack the capacity to fully understand a research question. In the assessment, the questions were written as simply as possible. However some of them could still have interpreted their own things.
Abstract concepts are difficult to understand, and they require people of higher than average intelligence to answer them correctly. If the research was only measuring concrete things like the cost of buying something, then chances are that fewer threats to the validity and reliability of the research would have arisen.
Additionally, since the research contained multiple choices, the rating involved could have affected the outcomes. Analysts have found that some people are in-between raters while others are extreme raters. The latter tend to choose mostly the best and the worst scores in a rating scale.
Conversely, in-between raters tend to select values that are in the middle. As a result, the responses in the questionnaire will be affected by one’s rating behavior. In this case, the study that was conducted may be questionable owing to this inclination.
Response biases also arise when a research pegs on self report. Some individuals may select certain answers even when they have observed evidence to the contrary. In these scenarios, many of them may choose yes regardless of only experiencing such occurrences once.
Alternatively, others may choose no even after encountering several experiences to the contrary. In this regard, the research will have to use their responses even when a number of them are simply demonstrating their response biases (Miller & Wright, 1995).
Some challenges could arise when the participants simply get tired of answering the questions or they lack the patience to read a question properly. Impatient people may not notice minor details that change the essence of a questionnaire dramatically.
Some may not understand a question and may lack the confidence to ask the researcher for clarification. In this case, they may provide answers only in accordance to their closest interpretation of the question, even if it is not accurate. Therefore, the answers they give could threaten the validity and reliability of the research.
Attrition or parental withdrawal
Attrition is a major threat to internal validity in this research. In this regard, some of the participants did not want to participate in the study. They withdrew from the analysis because of a number of reasons. Some of them felt that the questionnaires were boring so they did not respond to a number of the questions.
Additionally, certain subjects felt that the research would take too long for the research to be completed. Consequently, they apologized for their inability to participate in it and withdrew. Others did not see how the research would benefit them so they felt that it was not worthwhile to sacrifice their time.
Certain subjects did not justify their withdrawal. It is likely that some of them fell ill or had to leave town. Alternatively some may feel that the tasks involved in the research were arduous for them (Fan et. al., 2006).
Attrition can become a threat to validity of the research if the people who withdraw from the research possess characteristics that directly affect the parameters under analysis in the research. Additionally, subject withdrawal is a problem if the participants who left the research were more than 20%.
However, if the number is less than this percentage, then chances are that attrition will not be a significant source of bias. In the research, the percentage of persons who left the study was less than this percentage so they did not substantially affect the outcome of the research.
Additionally, the persons who dropped from the study did not have such unique traits that their absence could cause the rest of the participants to become different from the original sample.
One must consider these attrition levels in subsequent parts of the research such as analysis and discussion. This research will correct the problem by statistical means in the research analysis.
Fan, X., Miller, B. C., Park, K., Winward, B. W., Christensen, M. & Grotevant, H. D. (2006). An exploratory study about inaccuracy and invalidity in adolescent self-report surveys. Field Methods,18, 223–244.
Miller, R. & Wright, D. (1995). Detecting and correcting attrition bias in longitudinal family research. Journal of Marriage and Family 57, 921-929. Print.