#Overview

The graphs aim to analyze the self reported data of students regarding their sense of Locus of Control about their performance on their tests. The cohorts used are 2021 Fall, 2022 Fall, 2023 Fall, 2023 Spring, 2024 Fall, 2026 Spring. 2026 Spring has been separated into two different cohorts since some data was asked post dense and some was asked post exam. Additionally there are two different grading scales that were used, most data was collected with an internally created one.

Locus of Control Across all Cohorts and Exams

Most of the data seems to be pointing to an overall high internal locus of control for more students. This data used the internal grading scale so it does not include data from 2024 fall nor 2026 spring post exam (exam 1) data. Students were asked to report what influenced their exam performance from 1 to 100, 1 being completely outside factors and 100 being completely internal factors.

Internal_LOC

Negatively skewed distribution

External_LOC

positively skewed distribution were a high peak at 0 external LOC

Locus of Control per Cohort

Distributions resemble main distribution but differences across cohorts are noticeable. 2024 data set is not available because it was measured using an ASQ questionnaire instead of the manatee measuring scale from 1 to 100. The data for 2026 spring in dark blue represents the data taken post dense (exam 1, 2, and 3). The data in dark orange for 2026 spring was recorded post exam (exam 2, 3, and 4)

Internal_LOC

External_LOC

ASQ Grading Scale

These measures only apply to the 2024 fall cohort and the data for exam 1 post exam of spring 2026. For these cohorts a set of 12 factors was given to students from which they had to pick the top three most relevant factors that influenced their performance and score in the test. They were then asked to through the use of a slidebar give the factor a score from 1 to 7 reflecting their control on said factor, the handles for the scale were; 1 = attribute circumstances to others completely, 7 = attribute circumstance to me completely. The factors are displayed on the left side along the axis.

ASQ_Factor_Frequency

Mean_score_per_factor

The bars have error bars that correspond to one standard deviation below and above the mean.

Overlapped Cohort Distributions

Density distributions of cohorts that used MANATEE grading scale with slide bar from 1 to 100.

External_LOC

Internal_LOC

Distributions per Exam

The density of each of the cohorts per exam are overlapped to show the breakdown of the distributions per exam. For some exams data from certain cohorts is not present because it was not collected.

Internal_LOC_Exam_1

Internal_LOC_Exam_2

Internal_LOC_Exam_3

Internal_LOC_Exam_4

Internal_LOC_Exam_5

External_LOC_Exam_1

External_LOC_Exam_2

External_LOC_Exam_3

External_LOC_Exam_4

External_LOC_Exam_5

Distribution per Cohort (split by exam)

The distribution for fall of 2024 was not present because it was measured an ASQ grading Scale, while this data is the one measured by the in house MANATEE scale.

2021_Fall_Internal

2022_Fall_Internal

2023_Fall_Internal

2023_Spring_Internal

2026_Spring_Internal

2021_Fall_External

2022_Fall_External

2023_Fall_External

2023_Spring_External

2026_Spring_External

Scatterplot between ILOC and ELOC

These scatter plots aim to show the relationship between the indicated ILOC and ELOC reported by students ### ELOC_v._ILOC(stat smooth) In this case every recording made by every student was added as a data point this includes a recordin per post dense, pre dense and per exam. Additionally a smooth line was added with ggplot2 to try and inicate a relationship between the intenal and external locus of control.

ELOC_v._ILOC(lm smooth)

In this plot a linear regression line was added to show the possible relationship between two variables.

ELOC_v._ILOC(density_and_lm)

Ths plot was done as a density lot showing with colors and zones where most of the data is collected, additionally the linear regression model with the formula have been added to the plot

ELOC_v._ILOC(axis)

For this density plot the line x = y was added to the graph. Anything above the line shows a higher internal locus of control than external, and anything below the line indicates a higher external than internal locus of control reported. ### ELOC_v._ILOC(split_exam) Scatter plots were split by exam to show any observable impact time or exam number have on the reported external and internal locus of control. Smooth linear models were also added

ELOC_v._ILOC(slopes_exam)

Separate linear regression models are shown to indicate the variance between them per exam

Scatter plot between ILOC and ELOC average

In these plots an average was obtained of the reported internal and external LOC across exams in order to have a single data point per person involved in the study. As it can be observed most of the data is still clustered towards the top left indicating a higher internal locus of control rather than external locus of control. ### ELOC_v._ILOC

ELOC_v._ILOC(smooth)

A smooth regression line was added to indicate a possible relationship. 

## <ggplot2::labels> List of 1
##  $ title: chr "ELOC v. ILOC average LOC per person "

ELOC_v._ILOC(density_and_lm)

A linear regression was added as well as the formula to this density plot to help with visualization

LOC Ratio Histogram Distributions

A few histograms are shown however, because of the variability among the calculated ratios (Internal LOC / External LOC) the data is harder to visualize and I believe the density scatter plot do a better job at showing the distribution of the data collected.The distributions were drawn using every single data imput available meaning that there are multiple data points per person.

Distribution

Distribution(0-1_logscale)

This histrogam focuses on the data condensed between 0 and 1 and the y axis is turned into a log scale. The distribution seems to be negatively skewed indicating that most of the data is close to 0 and a larger internal locus of control compared to an external locus of control.

Distribution(0-0.9999999)

The y axis for this distribution was limited between 0 and 0.9999999, a regular scale was used.

Distribution(1-100)

The y axis was limited to 1 and 100, still most of the data seems to be closer to 0 as the grah os heavily positively skewed, normal scale was used.

Distribution (sigmoid)

A Sigmoid formula was applied to the ata in an attempt to better visualize the data. The data is still heavily skewed in this case negatively.

LOC Ractio v. Grade, PE1, PE2, EVPE

Ratio is graphed alongside variables to show possible relationship. All of the plots also have a linear model regression line. All of the possible variables that the ratio is compared to was graphed both for every data imput that was obtained and also with one data imput per person using average of both the average of the ratio and also the average of the variable.

Ratio_v._Grade

Ratio(average)_v._Grade

### Ratio(average)_v._Grade(average)

Ratio_v._PE1

Ratio(average)_v._PE1

Ratio(average)_v._PE1(average)

Ratio_v._PE2

Ratio(average)_v._PE2

Ratio(average)_v._PE2(average)

Ratio_v._EVPE

Ratio(average)_v._EVPE

Ratio(average)_v._EVPE(average)

Delta PE1

This statistic was calculated by subtracting PE1s to find the difference on what their predictions on what their exam grade would be to look how consistent this predictions in performances are among the sample.

Exam2_-_Exam1

Exam3_-_Exam2

Exam4_-_Exam3