IB Psychology HL Internal Assessment Example

This comparison with the multi-store model is imprecise. The student should tighten the explanation by stating exactly what Levels of Processing claims and avoiding vague wording like “doesn’t enter linearly,” which weakens the conceptual accuracy of the introduction.

The aim is stated clearly and is directly tied to the Levels of Processing theory. This works well because it gives the investigation a focused purpose that readers can immediately connect to the later method and conclusion.

The theory is identified and connected to the investigation topic through the contrast between shallow and deep processing. This gives the reader the conceptual basis for the experiment and shows why question depth should affect recall.

The hypothesis section identifies both variables clearly and operationally. This is strong because it makes the intended cause-and-effect relationship testable rather than leaving the variables as abstract labels.

The null hypothesis is fully operationalized with the exact question types and the measurable outcome. This helps the reader see precisely what is being tested and supports a clear statistical decision later.

The student explains why the investigation matters in a real school context. Linking the theory to study strategies and memorization shows relevance rather than just naming an abstract psychology concept.

The alternative hypothesis matches the null in structure and uses the same operational definitions. This consistency is important because it makes the comparison between hypotheses mathematically and logically clean.

The research design is clearly identified, which gives the investigation a solid methodological starting point. Naming the design allows the reader to understand how participants were allocated and why certain controls matter.

It would be stronger to add the full procedure here, especially how the words were presented, how the question-response cycle worked, and when recall was measured. That extra detail would let the reader evaluate the design more precisely.

The student explains why the design is suitable, not just what it is. Linking independent measures to the avoidance of practice and fatigue effects shows good awareness of design validity.

The sampling method is stated directly and is appropriate for the context of a school-based study. This makes the recruitment process transparent and easy to evaluate.

The participant description is specific and relevant to the investigation. Age, grade, and language fluency are all important here because they affect whether participants can understand the word task and be compared fairly.

The control variables are not merely listed; this one is linked to fairness between groups. Standardising processing time strengthens internal validity because it reduces the chance that time-on-task, rather than question depth, explains the results.

The student should specify what was randomized: the order of words, the assignment of question type, or both. That detail would make the design easier to replicate and would strengthen the explanation of how bias was reduced.

The materials section shows a clear attempt to control presentation conditions and reduce bias. This is good methodological practice because it helps isolate the independent variable rather than letting differences in presentation affect recall.

The table directly presents the two groups and their summary statistics, which is exactly what the reader needs before inferential testing. Clear separation of mean and standard deviation makes the pattern easy to interpret.

The student correctly recognises that the dependent variable is numerical and can be treated as ratio data. This matters because the choice of statistical test depends on the measurement level.

The student could make this section stronger by briefly explaining why mean and standard deviation were chosen for these data, rather than simply defining descriptive statistics in general terms. That would improve the link between method and analysis.

The graph is well chosen because it compares the mean recall scores for the shallow and deep conditions in a way that directly addresses the hypothesis. The inclusion of labelled axes and error bars makes the figure useful for interpreting the data.

This interpretation is not valid as written. Overlapping error bars do not by themselves prove non-significance, especially when a Mann–Whitney test is later reported as significant. The student should avoid using error-bar overlap as a substitute for inferential testing.

The student uses the test result to reach a statistical decision, which is an important part of analysis. Comparing the observed $U$ value with the critical value shows the inferential logic behind rejecting the null hypothesis.

The test is named correctly, but the explanation of why it was used is inaccurate. A Mann–Whitney test compares two independent groups, not “means from the same population,” so the rationale should be rewritten to show accurate statistical understanding.

It would help to state the exact $p$-value, the test statistic, and the decision rule in a clean step-by-step format. That would make the inferential analysis more transparent and easier to verify.

The conclusion is directionally correct, but the justification should be tied more explicitly to the chosen critical value and decision threshold. The student should also avoid repeating the null hypothesis in full here, which makes the conclusion harder to read.

The discussion connects the findings back to the background theory, which is exactly what this criterion requires. The student goes beyond reporting numbers by explaining how the pattern supports deeper semantic processing.

The student should finish this point by explaining exactly how being an all-girls sample or omitting another aspect of the model limits the interpretation. A more precise limitation would show how far the findings can or cannot support the broader theory.

This claim is overstated. A narrow age range may make the sample more specific, but it does not automatically mean participants have similar memory abilities. The student should replace this with a more defensible explanation, such as increased consistency in developmental stage or school experience.

The evaluation usefully connects the design choice to a relevant strength of the investigation. Explaining that independent measures reduce order effects shows the student understands why the method matters for validity.

This is a relevant limitation strength because it is tied directly to the hypothesis-testing context. The student correctly links the design to reducing participant guessing and behaviour changes.

This limitation is too vague. The student should explain what form of researcher bias is meant, how it could influence scoring or administration, and why it matters for the present hypothesis.

The limitation is relevant and clearly explained. The student shows why convenience sampling affects representativeness, which is stronger than merely naming the weakness.

The conclusion is clearly tied back to the hypothesis and the theory. This is effective because it gives a direct final answer to the research question instead of ending with a vague summary.

This would be stronger if the student named a concrete safeguard, such as using blind marking, a scoring rubric, or a second rater. Specific procedures are more convincing than a general statement about being aware of bias.

The modification is clearly linked to the limitation of a narrow sample. That connection is important because it shows the student is not just listing improvements but responding to a specific weakness in the design.

This is a strong modification because it explains both the method and the benefit. Linking random sampling to reduced selection bias and better representativeness shows good evaluative reasoning.