Erratum: The (Spatial) Memory Game: Testing the Relationship Between Spatial Language, Object Knowledge, and Spatial Cognition

DOI:

10.3791/6480

⸱

September 16th, 2021

In This Article

Summary

$$\rightleftharpoonup{xx}$$ $$\longleftharp{xx}$$, $$\longrightharp{xx}$$,

Use multilevel modelling (MLM) for analyses: An erratum was issued for: The (Spatial) Memory Game: Testing the Relationship Between Spatial Language, Object Knowledge, and Spatial Cognition. The Protocol was updated.

Abstract

$$\rightleftharpoonup{xx}$$ $$\longleftharp{xx}$$, $$\longrightharp{xx}$$,

Use multilevel modelling (MLM) for analyses: An erratum was issued for: The (Spatial) Memory Game: Testing the Relationship Between Spatial Language, Object Knowledge, and Spatial Cognition. The Protocol was updated.

Section 4 of the Protocol was updated from:

4. Analysis

  1. Language version (section 3.1)
    1. Calculate percentages of demonstrative use for each combination of variables.
    2. Analyze data with a mixed ANOVA, using within (e.g., distance, ownership, familiarity, visibility, language at instruction, position of a conspecific) and between (e.g., gender) variables1,13.
  2. Memory version (section 3.2):
    1. Calculate the absolute value of the difference between the actual location and the recalled location (in mm, negative values indicate the object was recalled to be closer) for all trials, then average this difference over each cell of the design.
    2. Analyze data in a mixed ANOVA, using within (e.g., distance, ownership, familiarity, visibility, language at instruction, position of a conspecific) and between (e.g., gender) variables13,14.

to:

4. Analysis using Multilevel Modelling (MLM)

  1. Identify potential zero-values in individual cells of the design, as these may lead to statistical separation6,7
  2. Use the Satterthwaite approximation8,9 to calculate effective degrees of freedom in the MLM.
  3. Use a backwards stepwise procedure for interaction terms to eliminate non-significant, highest order interactions:
    1. Retain all lower-order interactions if a related higher-order interaction is significant.
    2. Double check the changes in the model via the classification table, coefficients, and main effects with each iteration.
    3. Use the AIC to determine the efficiency of the model relative to the previous iteration10.
  4. Repeat Step 3 until reaching a model including all main effects, all significant (higher order) interactions (if any), and related lower order interactions (whether significant or not).

Update: Please find more information of this updated analysis replacing Section 4 of this publication at doi: https://osf.io/wnj2v/

Protocol

$$\rightleftharpoonup{xx}$$ $$\longleftharp{xx}$$, $$\longrightharp{xx}$$,

Use multilevel modelling (MLM) for analyses: An erratum was issued for: The (Spatial) Memory Game: Testing the Relationship Between Spatial Language, Object Knowledge, and Spatial Cognition. The Protocol was updated.

Section 4 of the Protocol was updated from:

4. Analysis

  1. Language version (section 3.1)
    1. Calculate percentages of demonstrative use for each combination of variables.
    2. Analyze data with a mixed ANOVA, using within (e.g., distance, ownership, familiarity, visibility, language at instruction, position of a conspecific) and between (e.g., gender) variables1,13.
  2. Memory version (section 3.2):
    1. Calculate the absolute value of the difference between the actual location and the recalled location (in mm, negative values indicate the object was recalled to be closer) for all trials, then average this difference over each cell of the design.
    2. Analyze data in a mixed ANOVA, using within (e.g., distance, ownership, familiarity, visibility, language at instruction, position of a conspecific) and between (e.g., gender) variables13,14.

to:

4. Analysis using Multilevel Modelling (MLM)

  1. Identify potential zero-values in individual cells of the design, as these may lead to statistical separation6,7
  2. Use the Satterthwaite approximation8,9 to calculate effective degrees of freedom in the MLM.
  3. Use a backwards stepwise procedure for interaction terms to eliminate non-significant, highest order interactions:
    1. Retain all lower-order interactions if a related higher-order interaction is significant.
    2. Double check the changes in the model via the classification table, coefficients, and main effects with each iteration.
    3. Use the AIC to determine the efficiency of the model relative to the previous iteration10.
  4. Repeat Step 3 until reaching a model including all main effects, all significant (higher order) interactions (if any), and related lower order interactions (whether significant or not).

Update: Please find more information of this updated analysis replacing Section 4 of this publication at doi: https://osf.io/wnj2v/

Disclosures

$$\rightleftharpoonup{xx}$$ $$\longleftharp{xx}$$, $$\longrightharp{xx}$$,
No conflicts of interest declared.

Reprints and Permissions

Request permission to reuse the text or figures of this JoVE article

Request Permission

Tags

Multilevel ModellingSpatial LanguageObject KnowledgeSpatial CognitionMixed ANOVASatterthwaite ApproximationBackwards Stepwise ProcedureAIC Model EfficiencyZero Value IdentificationClassification Table Analysis