13.3
Introduction au test des signes
Le test des signes est un outil important dans les statistiques non paramétriques, offrant une méthode simple mais efficace pour analyser des paires a…
Le test des signes est une méthode non paramétrique permettant d’évaluer les affirmations concernant des données aléatoires simples provenant de paires appariées, de données nominales ou d’affirmations concernant la médiane de la population.
Il transforme les données en signes positifs et négatifs sur la base d’hypothèses prédéterminées et évalue si la différence dans le nombre total de chaque signe est statistiquement significative.
L’hypothèse nulle du test du signe propose que les caractéristiques de la population s’alignent sur les affirmations, tandis que l’hypothèse alternative suggère le contraire.
Pour les ensembles de données où le nombre total de signes ne dépasse pas 25, la statistique de test, notée x, correspond à la quantité du signe le moins fréquent.
Dans les cas où le total dépasse 25, la statistique de test, représentée par z, est calculée.
Des tables spécifiques sont utilisées pour déterminer les valeurs critiques. L’hypothèse nulle est rejetée si la valeur statistique du test est inférieure ou égale à la valeur critique. Sinon, il y a échec du rejet de l’hypothèse nulle.
View the full transcript and gain access to JoVE Core videos
Q1: What is the sign test and when should you use it?
The sign test is a nonparametric method for evaluating claims about matched pairs, nominal data, or population median assertions. It transforms data into positive and negative signs based on predetermined assumptions and assesses whether the difference in sign counts is statistically significant. This test is particularly valuable when data does not conform to normal distribution requirements.
Q2: How do you assign signs to data in a sign test?
For each pair of observations, compare their values: assign a plus sign if sample A exceeds sample B, a minus sign if sample A is less than sample B, and discard the pair if values are equal. Count the resulting positive and negative signs to proceed with the test. This straightforward approach avoids assumptions about data distribution.
Q3: What is the null hypothesis in a sign test?
The null hypothesis proposes that population characteristics align with the claims being tested, such as no difference in medians between two populations. The alternative hypothesis suggests the opposite. If a predominance of one sign over the other emerges, it may indicate a statistically significant effect contradicting the null hypothesis.
Q4: How does sample size affect the sign test statistic calculation?
For datasets with 25 or fewer observations, the test statistic (x) represents the count of the less frequent sign. For larger datasets exceeding 25 observations, a z-score is computed instead. Both approaches facilitate comparison against critical values from statistical tables to determine significance.
Q5: When do you reject the null hypothesis in a sign test?
The null hypothesis is rejected if the test statistic value is less than or equal to the critical value obtained from statistical tables. If the test statistic exceeds the critical value, there is insufficient evidence to reject the null hypothesis, indicating no statistically significant difference.
Q6: What is a practical example of sign test application?
Researchers might employ the sign test to evaluate pre- and post-treatment effects in a medical study, determining whether treatment correlates with improvement (positive sign) or deterioration (negative sign) in patient outcomes. This application demonstrates how the sign test assesses directional changes without requiring distributional assumptions.
Q7: How does the sign test compare to other nonparametric alternatives?
The sign test offers a simpler approach than alternatives like the wilcoxon signed ranks test for matched pairs, which incorporates magnitude information. The sign test focuses solely on direction of change, making it more robust for ordinal data or when exact values are unreliable, though potentially less powerful statistically.