Criterion E: Analysis and conclusion (6)

Pictureimage from /stock.adobe.com
This criterion assesses the extent to which the student has effectively communicated and processed the
data in ways that are relevant to the research question. The student should utilize techniques associated
with the appropriate experimental or social science method of inquiry.

If there is insufficient data then any treatment will be superficial. You need to recognize the potential for such a lack and revisit the method before you arrive at the data collection or analysis. Alternatively, a lack of primary data could be supplemented by the use of secondary data from data banks or simulations to provide sufficient material for analysis.
Clarification for Treatment Data

  • Minor errors (those that do not affect the conclusion) should not prevent a report from achieving full marks for the criteria/performance level.
  • Data can be primary or secondary, and qualitative or quantitative.
  • Clear means that the presentation or method of processing can be understood easily, including appropriate details such as the labelling of graphs and tables or the use of units, decimal places and significant figures, where appropriate.
  • The raw data presented might be a sample if there is a large amount, that is, survey results or data logging; the remaining data can be included within an appendix.
Guidelines for Data Analysis
  • Discuss Trends and Patterns: Identify and explain any observed trends, patterns, or correlations in the data, referencing specific figures (e.g., “According to Figure 1…”).
  • Interpret Data for Conclusions: Interpret the data meaningfully to draw relevant conclusions.
  • Statistical Significance: If statistical tests were used, interpret the significance of their results clearly within the context of your investigation.
  • Address Uncertainties: Discuss the impact of uncertainties on the data and conclusions, noting any factors that may limit the confidence in the results.
  • Evaluate Bias, Reliability, and Validity: Assess potential biases, reliability, and validity in your data, especially for surveys. Recognize any weaknesses in your data and comment on how these may affect results.
  • Reference Scientific Literature: Where possible, compare findings to scientific literature or accepted ranges to support or challenge your data’s validity.
  • Clarify Relationships: Accurately identify relationships (e.g., distinguishing a negative correlation from an inverse relationship) to avoid misinterpretation.
Guidelines for Considering Uncertainties in Data Analysis
  • Acknowledge Uncertainties: Recognize that uncertainties are inherent in any investigation. Address as many as possible in your method, and consider their impact on your results.
  • Evaluate Sources of Uncertainty: Identify where uncertainties arise in your experiment, and discuss their potential effects on data interpretation. This can lead to insights for improving the method or suggesting further questions.
  • Measure Variation: Use measures like range and standard deviation to assess result reliability. Larger sample sizes typically reduce uncertainty, enhancing accuracy and allowing for more precise conclusions.
  • Use Statistical Indicators:
    • Standard Error of the Mean: Helps gauge accuracy, especially with larger sample sizes, and supports establishing 95% confidence limits.
    • Coefficient of Determination (R²): Shows how well a trend line fits the data, indicating reliability in trend analysis.
  • Interpret p-Values: For statistical significance tests, a p-value under 0.05 (5%) indicates significance, suggesting results are unlikely due to chance. Interpret these results carefully in context, knowing that p-values have limitations but remain useful in rejecting or supporting hypotheses.
Guidelines for Drawing Relevant Conclusions
  • Address the Research Question: Ensure your conclusion directly responds to the research question.
  • Reference Hypotheses: If a hypothesis was stated, indicate whether the data supports or refutes it.
  • Base Conclusions on Data: Draw conclusions from the evidence, not assumptions. Recognize that variability in data often leads to tentative conclusions that may highlight patterns or trends rather than causation.
  • State Relevance and Limitations: Clearly state if your data sufficiently addresses the research question. If results are inconclusive, reflect this in the conclusion.
  • Limit Scope of Conclusions: Avoid overgeneralizing based on models or simulations. Your conclusion should be specific to your data and may not apply broadly to the real-world phenomenon being modeled.
​Where possible, the variability should be demonstrated and explained, and its impact on the conclusion fully acknowledged. Please note, by “conclusion”, is meant a deduction based on the direct interpretation of the data such as “What does the graph show?” or “Does any statistical test used support the conclusion?” Any overview of the data in the light of the broader context will be assessed in the criterion for discussion and evaluation.

Conclusion:
  • The conclusion given is correct and clearly supported by the interpretation of the data.
  • Key data from the analysis is given and trends in the data are discussed.
  • The extent to which the hypothesis is supported by the data is explained (avoiding “proves”).
  • The level of support (strong, weak, none or inconclusive) for the hypothesis/ conclusion is identified, correct and justified.

The conclusion
  • starts with one (or more) paragraphs in which you draw conclusions from results, and state whether or not the conclusions support the hypothesis .  
  • clearly related to the research question and the purpose of the experiment.  
  • uses the expressions ‘confirmed by the data’ or ‘refuted by the data’ rather than ‘right,’ ‘wrong,’ or 'proven.'
  • provides a brief explanation as to how you came to the conclusion from the results.  In other words, sum up the evidence and explain observations, trends or patterns revealed by the data. 

Scientific Context:
  • Scientific explanation for the results is described.
  • Comparison is made with published data and theoretical texts (with citations).

The results of the experiment should
  • be explained using accurate and relevant science.  
  • should compare the results of your investigation with what would be expected; reference published data or theoretical texts.
  • compare the conclusions with published research or with the general scientific consensus among scientists about the research question.  Do your conclusions conform to the consensus or are they unexpected?

 It is not necessary to find an exact same investigation with the exact same results, it is possible to compare findings with another investigation that is different but with results that either confirm or refute those of the student's investigation.