k-Nearest Neighbors for Classifying Students' Test Anxiety Level from Test Scores and Study Hours

Project info

• Field of studyPsychology
• LevelUndergraduate thesis (S1)
• Method typeClassification
• Estimated build time± 3 working days
• Toolsscikit-learn KNeighborsClassifier, cross_val_score, StandardScaler
• PriceUS$19 - US$113

Research goal

Classify test anxiety level with k-Nearest Neighbors based on test score and study hours and choose the best k.

Variables

Example result

• 0.939Accuracy
• 0.889Precision
• 1Recall
• 0.941F1-score
• 0.974ROC AUC
• 3Best k

The k-NN model classifies Test anxiety level (low/high) from Test score and Study hours with test accuracy 0.939, F1 0.941 and AUC 0.974. Best k = 3.

Computed from illustrative synthetic data, not from a real study.

About k-Nearest Neighbors

k-Nearest Neighbors classifies a point by the majority class among its k nearest neighbours. It suits as a simple baseline for small to medium data with irregular class boundaries.

Steps

1. 1Standardise features because it is distance based
2. 2Split into training and test sets
3. 3Choose the best k by cross-validation
4. 4Train KNeighborsClassifier with the chosen k
5. 5Evaluate accuracy, F1, and the confusion matrix
6. 6Plot accuracy against k

Assumptions to check

• Features are comparably scaled and relevant
• The distance metric fits the data
• High dimensionality weakens the meaning of distance (curse of dimensionality)

Reading and reporting the results

Report the chosen k with the reason for choosing it, and the test metrics. Note that prediction is slow on large data because the whole training set is stored.

Example Python code

A short starting script with this project's variable names. The delivered version is adapted to your data and fully commented.

import numpy as np
import matplotlib.pyplot as plt
from sklearn.neighbors import KNeighborsClassifier
from sklearn.preprocessing import StandardScaler
from sklearn.pipeline import make_pipeline
from sklearn.model_selection import cross_val_score, train_test_split
from sklearn.metrics import accuracy_score, f1_score

rng = np.random.default_rng(51)
n = 300
test_score = np.r_[rng.normal(0, 1, n // 2), rng.normal(2, 1, n // 2)]
study_hours = np.r_[rng.normal(0, 1, n // 2), rng.normal(2, 1, n // 2)]
test_anxiety_level = np.r_[np.zeros(n // 2), np.ones(n // 2)].astype(int)

X = np.column_stack([test_score, study_hours])
X_tr, X_te, y_tr, y_te = train_test_split(X, test_anxiety_level, test_size=0.25, stratify=test_anxiety_level, random_state=51)
ks = range(1, 26, 2)
cv = [cross_val_score(make_pipeline(StandardScaler(), KNeighborsClassifier(k)), X_tr, y_tr, cv=5).mean()
      for k in ks]
best_k = list(ks)[int(np.argmax(cv))]
clf = make_pipeline(StandardScaler(), KNeighborsClassifier(best_k)).fit(X_tr, y_tr)
pred = clf.predict(X_te)
print("Best k:", best_k)
print("Accuracy:", round(accuracy_score(y_te, pred), 3), "F1:", round(f1_score(y_te, pred), 3))

plt.plot(list(ks), cv, marker="o")
plt.xlabel("k"); plt.ylabel("CV accuracy")
plt.show()

Packages and prices

Choose how much help you need. Each higher package includes everything in the lower ones. Prices exclude 11% VAT.

• Payment: 50% down payment before work starts, balance after the deliverables are handed over
• Tax: Prices exclude 11% VAT
• Data and academic integrity: Results are computed from your own research data; you write and stand behind your manuscript, we help with the analysis and your understanding of it
• Revisions: Up to 2 revisions are included in the price (Package 3: 3 rounds)

Frequently asked questions