Sample Size for Multiple Regression: Obtaining Regression Coefficients That Are Accurate, Not Simply Significant.

Kelley, Ken; Maxwell, Scott E.

doi:10.1037/1082-989x.8.3.305

Cited by 262 publications

(194 citation statements)

References 50 publications

Supporting

Mentioning

189

Contrasting

Unclassified

Order By: Relevance

“…Thus, there are different basic rules that establish the number of participants needed to predict. If we take into account the considerations in Green (1991) and Kelley & Maxwell (2003) of a minimum of 10 to 15 observations per predictor, we can affirm that 65 participants for an analysis of two predictors per model provide the minimum required to implement a high-quality multiple linear regression analysis.…”

Section: Discussionmentioning

confidence: 87%

Cognitive presence through social and teaching presence in communities of inquiry: A correlational–predictive study

Gutiérrez-Santiuste

Rodríguez-Sabiote

Arrufat

2015

AJET

View full text Add to dashboard Cite

This study focuses on the relationships established between the elements that compose the community of inquiry (CoI): cognitive, social, and teaching presence. Using three questionnaires, we analyse the students' perception of synchronous and asynchronous virtual in text-based communication (chats, forums and emails). Starting from the high correlations found between the three elements, we perform a multiple linear regression analysis. The findings show that relationships can be established in the model in which the cognitive elements are strongly predicted, to a greater extent by social presence than by teaching presence. In the forums the cognitive presence is better explained by other presences than in chats and emails. The results reveal the need to analyse the three kinds of presence jointly, assessing the impact of each on student learning. We also determine that the instructor benefits from knowing which tool is more valid for the learning objectives.

show abstract

Section: Discussionmentioning

confidence: 87%

Cognitive presence through social and teaching presence in communities of inquiry: A correlational–predictive study

Gutiérrez-Santiuste

Rodríguez-Sabiote

Arrufat

2015

AJET

View full text Add to dashboard Cite

show abstract

“…Several authors, however, have argued that any field of science should move from binary decisions derived from the NHST procedure towards giving a more precise point estimate of the magnitude of an effect (Edwards & Berry, 2010;Kelley & Maxwell, 2003).…”

Section: Introductionmentioning

confidence: 99%

At what sample size do correlations stabilize?

Schönbrodt

Perugini

2013

Journal of Research in Personality

1,821

1,265

View full text Add to dashboard Cite

Sample correlations converge to the population value with increasing sample size, but the estimates are often inaccurate in small samples. In this report we use Monte-Carlo simulations to determine the critical sample size from which on the magnitude of a correlation can be expected to be stable. The necessary sample size to achieve stable estimates for correlations depends on the effect size, the width of the corridor of stability (i.e., a corridor around the true value where deviations are tolerated), and the requested confidence that the trajectory does not leave this corridor any more. Results indicate that in typical scenarios the sample size should approach 250 for stable estimates.

show abstract

“…The reduced final model is built with the use of 120 observations which, for such a linearized one variable model, constitute a sufficient sample size (Kelley and Maxwell 2003;Soper 2015). It is confirmed by a relatively high value of coefficient of determination (r 2 = 0.84).…”

Section: Uhi On the Basis Of Regression Modelsmentioning

confidence: 99%

The urban heat island in the city of Poznań as derived from Landsat 5 TM

Majkowska

Kolendowicz

Półrolniczak

et al. 2016

Theor Appl Climatol

View full text Add to dashboard Cite

To study urban heat island (UHI), Landsat 5 TM data and in situ measurements of air temperature from nine points in Poznań (Poland) for the period June 2008-May 2013 were used. Based on data from measurement points located in different types of land use, the surface urban heat island (SUHI) maps were created. All available and qualitycontrolled Landsat 5 TM images from 15 unique days were used to obtain the characteristics of land surface temperature (LST) and UHI intensity. In addition, spatial analysis of UHI was conducted on the basis of Corine Land Cover 2006 dataset. In situ measurements at a height of 2 m above ground level show that the UHI is a common occurrence in Poznań with a mean annual intensity of 1.0°C. The UHI intensity is greater during the warm half of the year. Moreover, results based on the remote sensing data and the Corine Land Cover 2006 indicate that the highest value of the mean LST anomalies (3.4°C) is attained by the continuous urban fabric, while the lowest value occurs within the broad-leaved forests (−3.1°C). To re-count from LST to the air temperature at a height of 2 m above ground level (T agl ), linear and non-linear regression models were created. For both models, coefficients of determination equal about 0.80, with slightly higher value for the non-linear approach, which was applied to estimate the T agl spatial variability over the city of Poznań.

show abstract

Sample Size for Multiple Regression: Obtaining Regression Coefficients That Are Accurate, Not Simply Significant.

Cited by 262 publications

References 50 publications

Cognitive presence through social and teaching presence in communities of inquiry: A correlational–predictive study

Cognitive presence through social and teaching presence in communities of inquiry: A correlational–predictive study

At what sample size do correlations stabilize?

The urban heat island in the city of Poznań as derived from Landsat 5 TM

Contact Info

Product

Resources

About