Table 6. Linear and logistic regressions...

Table 6.

Linear and logistic regressions testing the dependency of the companies’ posting behavior

Dimension	Predictor (Independent variable)	Company posting (dependent variable)
		No. of words Model D: Linear Regression	Use of picture Model E: Log. Regression	Company-relevant content Model F: Log. Regression
		ß (stand.)	Exp. (ß)	Exp. (ß)
Exogenous factors
Point in time	Time of the day: Leisure time	−0.080	0.672	1.530
	Day of the week	REFERENCE: Monday
	Tuesday	−0.008	1.492	1.036
	Wednesday	−0.082	0.846	2.257^a
	Thursday	−0.118	0.424^a	1.446
	Friday	−0.032	0.924	0.778
	Saturday	0.077	1.605	0.514
	Sunday	−0.139*	0.464	0.442
	Holiday month	0.064	1.051	0.210***
	Public holidays	−0.078	0.363	2.108**
Weather
	Temperature ^b	−0.205*	0.961^a/e	1.181
	Precipitation ^c	0.029	0.996	1.100 *^/e
	Humidity ^b	0.001	0.988	0.990
	Barometric Pressure ^b	−0.123*	0.956^/e*	1.025^/e*
	Wind velocity ^b	0.007	0.959^/e	1.007
	sunshine ^b	−0.044	0.811	1.019
Model	r²	0.109	0.127	0.154

Dimension	Predictor (Independent variable)	Company posting (dependent variable)
		No.of wordsModel D:Linear Regression	Use ofpictureModel E:Log. Regression	Company-relevant contentModel F:Log. Regression
		ß (stand.)	Exp. (ß)	Exp. (ß)
Exogenous factors
Point in time	Time of the day: Leisure time	−0.080	0.672	1.530
	Day of the week	REFERENCE: Monday
	Tuesday	−0.008	1.492	1.036
	Wednesday	−0.082	0.846	2.257^a
	Thursday	−0.118	0.424^a	1.446
	Friday	−0.032	0.924	0.778
	Saturday	0.077	1.605	0.514
	Sunday	−0.139*	0.464	0.442
	Holiday month	0.064	1.051	0.210***
	Public holidays	−0.078	0.363	2.108**
Weather
	Temperature ^b	−0.205*	0.961^a/e	1.181
	Precipitation ^c	0.029	0.996	1.100 ***^/e
	Humidity ^b	0.001	0.988	0.990
	Barometric Pressure ^b	−0.123*	0.956*^/e	1.025*^/e
	Wind velocity ^b	0.007	0.959**^/e	1.007
	sunshine ^b	−0.044	0.811	1.019
Model	r²	0.109	0.127	0.154

Notes:

Significance^d: italic letter = ^a0.1, *0.05, **0.01, ***0.001; Logistic Regression Nagelkerke r². ^bAt the time of posting. ^cOn the day of posting. ^dDue to the explorative character of this study and the limited sample of 321 postings, an expanded approach of four significance levels is applied (p < 0.001; p < 0.01; p < 0.05; p < 0.1). ^eProbability of an image or company-relevant content in a posting when changing the item indicators by one unit. For example, an increase from 5°C to 6°C

Source: Own compilation

[ViewLarge]

Sharing Unavailable