Universities Selection Model

1 collaborator

Razan Nofal (Author)

WHAT IS IT?

This model explores and replicates students selection of Jordanian universities in cities and urban areas. There are 30 university with different ranks in Jordan distributed randomly across different areas. The model shows the distribution of students in different universities based on the university reputation, their family income and their targets for each semester(tick). The model will stop automatically after 50 semester wich is 4 monsth.

If the student family income is high, the student will select and enroll in universities with high reputation and short distance otherwise they will be enrolled in moderate and low reputation universities.

The study replicates and simulates how students choose the appropriate universities based on some factors; family income, university reputation and area. There is a high number of students and universities options to choose each period. Most universities in cities have a high rank however the urban areas have small number of universities with moderate to low rank.

This model is useful for the educational and social departments to predict the number of students enrolled in the universities based on rank to improve the universities to attract all types of students based on their targets.

HOW IT WORKS

The model contains 2 areas (patches); Cities and urban areas in Jordan. The cities part is large with dark blue and the other part is the urban areas with green color. The model includes 2 agents; students (persons) and universities. Universities agent replicate the number of universities in Jordan (30 university). There is a slider to select the number of universities in the model, each university has a reputation variable. The reputation variable is very important factor for students to select their target university; there are some universities with high reputation factor in cities. The universities distributed randomly across cities and urban areas. the reputation depends on the value of rank slider and the colors reflected the rank; if the rank between 0 and 9 the university will have a high reputation with white color. If it's between 10 and 19, the reputation is moderate with yellow color. If it's between 19 and 30, it's a low rank with orange color. There is a variable for enrollment to calculate the enrolled student for each semester. Students (Persons) agent, the model represents a sample of 1500 students selected randomly and move between university based on their family income, university reputation and area. There are sliders to select the number of students. The high-income level in Jordan considered in the model as well. We can control the income level and rank to show how many students distributed across different universities and areas. The students move between different universities to reach their target. There is a target variable that defines the target of each student and movement based on his requirements and income.

The model will stop automatically after 50 semester (ticks). The user can continue the running of model.

There are monitors and plots shows the number of high, moderate and low reputation universities. There are 2 plots as well to show the enrolled students for each university by rank and the number of students enrolled in cities vs urban areas.

HOW TO USE IT

There are 4 sliders (the Model Parameters) to select the value of students, universities, university rank and high-income level.

The user can select: - The number of students to be considered in the model - The number of universities from 1-30 - The high-income level as we calculate the high family income based on that - The rank if universities 1-30.

The family income is considered high based on the income level selected in the model and the rank of universities is affecting the reputation calculation. The user can select go to run the model. There are multiple monitors to show the number of high, moderate and low rank universities. others show the number of enrolled students by each ranked universities and the percentage of enrolled student to all students. Furthermore, the model shows 2 plots for the number of students enrolled in universities in urban and cities areas and ranked universities.

THINGS TO NOTICE

The model is replicating a sample of students apply and select one of Jordanian universities in each semester based on their targets considering the distance to universities as well. The model is not reflecting real values, its reflecting values based on my understanding of student selection criteria and university ranks and expected values however the number of universities and areas are real data.

The high income factor in Jordan is approximate and not reflecting a real value as the high-income level is vary based on multiple factors.

EXTENDING THE MODEL

The model built is simple to reflect the students’ selection and distribution across different universities in Jordan. There are multiple factors that can be considered in future models such as student GPA, student target major, universities acceptance criteria. The model can be extended to classify the universities by private and public as it’s one of the factors that affecting the selection and students target..

NETLOGO FEATURES

Breeds: the use of breeds to model two different agents; students(persons) and universities.

move-to: to move agent to the targeted university.

Distance: reports the shortest distance between the agent and another provided agent (students and universities)

one-of: reports one randomly selected member from agent set.

RELATED MODELS

Models in Netlogo library:

SchoolChoiceABM created by Diego A. Díaz. Move Towards Target Example.

Citation: Diego A. Díaz, Ana María Jiménez & Cristián Larroulet (2019) An agent-based model of school choice with information asymmetries, Journal of Simulation

Wilensky, U. (1999). NetLogo. http://ccl.northwestern.edu/netlogo/. Center for Connected Learning and Computer-Based Modeling

CREDITS AND REFERENCES

Comments and Questions

Please start the discussion about this model! (You'll first need to log in.)

Click to Run Model

breed [persons person]
breed [universities university]

persons-own [target familyincome enrolled?]
universities-own [highreputation? reputation enrolledstudents]


; reset button

to setup
  clear-all
  setup-patch
  setup-university
  setup-person
  reset-ticks
end 


; setup the patch colors to city and urban areas

to setup-patch
  ask patches with [pxcor < -6  AND pycor >= -16]
  [
    set pcolor green
  ]
  ask patches with [pxcor >= -6 AND pycor <= 16]
  [ set pcolor cyan - 1
  ]
end 


; set up the universities based on reputation - it has different colors based on reputation

to setup-university
  create-universities number-of-universities  [fd max-pxcor setxy random-xcor random-ycor]
    ; set the universities shape and default color
     ask universities [ set shape "house two story"  set size  2.6
                   set color white

     ; high reputation with white color, moderate reputation with yellow and low reputation with orange
     ; reputation is based on rank of 30 university - low rank means high reputation and vice versa
     set reputation random 100 * university-rank / 100
     if (reputation >= 0 and reputation < 10 ) [set highreputation? true]
        ifelse (reputation >= 10 and reputation < 19) [set highreputation? true set color yellow][
        ifelse (reputation >= 19) [set highreputation? false set color orange + 1 ][set highreputation? true]]
        set enrolledstudents 0]
end 

; setup students color and shape and the target

to setup-person
  create-persons number-of-students
  [setxy random-xcor random-ycor set shape "person" set size 1
    set familyincome one-of (range 500 6000)
    set color blue - 2
               set target 1
    set target one-of universities
               face target

  ]
  reset-ticks
end 


; movement of students between universities to select their target based on distance as well

to move-students
 ask persons [
   if distance target = 0
      [ set target one-of universities
        face target ]

    ifelse distance target < 1
    [
      move-to target
      ]
      [ fd 1 ]

  ]
end 

; pick the target university based on family income, rank and distance, the university will be selected.

to select-university

        ifelse (familyincome > income-level and color =  white )
    [
set target one-of universities with [[familyincome] of myself] with-max [reputation]
    ]

  [
    ifelse (familyincome < income-level and  color =  yellow) [
      set target min-one-of universities with [[familyincome] of myself - (distance myself) / 100 > 0] with-max [reputation] [distance myself]]


    [ifelse (familyincome < income-level and color =  brown)
      [set target min-one-of universities with [[familyincome] of myself - (distance myself) / 100 > 0] with-min [reputation] [distance myself]]

       [set target min-one-of universities
      with [[familyincome] of myself - (distance myself) / 100 > 0] [distance myself] ]
  ]]

  if target != nobody
  [ move-students]
   set enrolled? TRUE
end 


; run the model by calling select university and move students to target

to go
  ask persons [
  select-university
  move-students]
  ask universities [ calculate-enrolledstudents ]
  tick
  if (ticks = 50)
  [stop]
end 

; proc to calculate the count of enrolled students

to calculate-enrolledstudents

  if (enrolledstudents < number-of-students)
  [set enrolledstudents enrolledstudents + 1]
end

There is only one version of this model, created almost 3 years ago by Razan Nofal.

Attached files

File	Type	Description	Last updated
Universities Selection Model.png	preview	Preview for 'Universities Selection Model'	almost 3 years ago, by Razan Nofal	Download

This model does not have any ancestors.

This model does not have any descendants.

NetLogo