CS-411 : Digital Education & Learning Analytics - ppt download

$ 11.00 · 4.9 (662) · In stock

(Orchestration Graphs) Analysis Learning Analytics Experimental methods Object Model Instruction Design (Orchestration Graphs) Synthesis
Chapter 8: Measuring learning effects.
Analysis. Learning Analytics. Experimental methods. Object. Model. Instruction Design. (Orchestration Graphs) Synthesis.
Dependent Independent Variable Variable Solo / Team Test Score
Factor. Independent variable. Modality 1. Modality 2. Condition. with. without. Experimental Group. Control Group.
Research Question. Is it more effective to watch MOOCs. individually or in small teams Research Sub Questions. It depends on the size of the group. It depends on the school culture. It depends on difficulty of the video. It depends how well the group members know each other. It depends …
Factor 1. Modality 1.1. Modality 1.2. Modality 2.1. Group A. Group E. Factor 2. Modality 2.2. Group B. Group D. Modality 2.3. Group C. Group F.
on the dependent variable. depends upon. the other independent variable.
Experiment Biases. Were the groups really equivalent at the beginning same age (in average)  Questionnaire. same gender ratio  Questionnaire. same socio-cultural level  Questionnaire. same school level  Recruitment. same background knowledge  Pre-test. same level for mutual knowledge (friends )  Recruitment. same level of intelligence  Pre-test. same spatial reasoning  Pre-test.
SOLUTIONS. (1) Control before forming groups.  When assigning subjects to conditions. distribute Equally among conditions: controlled variables. (2) Control after the experiment.  Verify that the groups do not differ significantly: randomized variables.  If the groups differ significantly, use covariate analysis. (3) Use repeated measures also called within subjects
The same subjects pass in both conditions. Experimental Plan: Within Subjects, 1 dimension. Factor. Modality 1. Modality 2. Subjects 1, 3, 5. First. Second. Subjects 2, 4, 6. Second. First. COUNTER-BALANCING. (avoiding the order effect) Pro: fewer subjects, samples are equivalent. Cons: complex order effects.
2. Why are they better Is it more effective to watch MOOCs. individually or in small teams
3. Why are they better Effects. Conditions. Independent Variables. Dependent Variables. Processes.
3. Why are they better Social facilitation Explanation effect More knowledge …. Effects. Conditions. Independent Variables. Dependent Variables. Processes. Process Variables.
3. Why are they better Verbal Interactions. Turn Taking mechanisms. Non-verbal interactions. Social relationships. Gaze Patterns. Deictics. …. Effects. Conditions. Independent Variables. Dependent Variables. Processes. Process Variables. Recorders. Eye trackers. Questionnaires (NASA TLX)
Example : The effect of persistency of information
Independent Variables. - Dependent Variables. Persistency of. information (‘vignettes’) - Score at post-test. + Process Variables. Frequency of content-rich interactions.
3. Why are they better Conditions. Processes. Effects. Independent Variables. Dependent Variables. Process Variables. Mediation Effect.
Is it more effective to watch MOOCs. individually or in small teams 3. Could this difference be explained by chance
Performance. Group 1. Group 2. Condition 1. Condition 2. Is this difference in group means due to sampling or is it the effects of condition changes If group 2 had been in condition 1 and group 1 in condition 2, would group 2 still get a higher performance
Score m = 37.2 m = 38.1 m = 2.8
Parobability Score m = 2.8 m = 37.2 m = 38.1 m = 61
The sample size reduces the probability to get by chance a sample mean that is far from the population means.
Dependent Variable (e.g. task performance) Dependent Variable (e.g. task performance) Condition A. Condition B. Condition A. Condition B.
Significant does not mean large !
Row in classroom. Teacher: A or B Learning style. IQ Test. Mental Rotation test. Motivation test. Level (low, medium, high) Opinions. Pre-test score. Post-test score. Learning gain. Response time. Balance of participation. Number of help requests. Total fixarion time on X. Gaze paths. Sequences of actions. ≠ types of variables. ≠ distributions. ≠ analyses. Nominal, discrete. Ordinal, ordered. Metric, continuous, normal ( ) Times series.
Ind: LearnStyle= Surface. Dep: Style = Leader. Ind: Motivation= Low. Dep: CodeQuality= Low. Ind: IQ= 142. Dep: Score= 23. Time. Series. Ind: {play pause back…} Dep: gaze {O1 O3 O1 O2…} Ind: Attention {3 3 5 …} Dep: {low low med …} Ind: HeartRate {60 90 …} Dep: RespTime { …} 2D & More. Gaze {(X1 Y2, α1, T1}, …. }
(Scale) Metric. (Continous) Categories. (Discrete) CHI-SQUARE. KRUSKALL. ANOVA. Independent Variable. Ordinal. (Scale) ANOVA. CHI-SQUARE. Supervised Learning. Metric. (Continous) Supervised Learning. REGRESSION.
Do students who watch the MOOC at the same time succeed better Solo / Team. VideoPlay (time, hour) Post-hoc. Analysis. Designed experiment. Test Score. Test Score.
Variable. Variable. Hidden Variable. causality.
Hypothesis Expected results ( A > B); an affirmation. Independent variables What one varies between the conditions (or Factor) Modality Value of a factor. Condition Set of (factor, modality) per group of subjects. Control group The reference against which one will compare. Dimension Number of factors. Dependent variables How does one measure the effects Controlled variables Things you try to keep constant or to randomize. Intermediate variables Explain the link from Independent to Dependent Variables. Significant difference Probably (<5%) not due to sampling error. Interaction effect The effect of one IV on the DV depends upon another IV. Between/Within subject Do subjects pass in one or several conditions Counterbalancing Inverting the order of conditions for within-subject plans.