I will buy virtual goods if I like them: a hybrid PLS-SEM-artificial neural network (ANN) analytical approach

Abstract

Despite the popularity of Free-to-Play (F2P) games in recent years, the motivations behind players’ intention to purchase virtual goods in F2P games still require further investigations. This study aims to address this dilemma by investigating the antecedents of functional, emotional, and social values in shaping the purchase intention of virtual goods in F2P games. Using purposive sampling, data were collected through a survey from 352 F2P game participants in the United States. A hybrid PLS-SEM-Artificial Neural Network (ANN) modeling approach was employed to examine the impact of these factors on the intention to purchase virtual goods. The results reveal that perceived value positively influences the purchase intention of virtual goods. The findings also show that functional, emotional, and social values significantly impact the perceived value and purchase intention of virtual goods. Further, perceived value mediates the relationship between quality, achievement, enjoyment, aesthetics, customization, self-presentation, and the intention to purchase virtual goods. The ANN results reveal that quality and social presence are the most critical factors since they achieve the greatest normalized importance ratio compared to the others. The model illustrated considerable explanatory evidence for purchase intention in the context of F2P games. Additionally, this research significantly strengthens the marketing literature by developing an understanding of the intention to buy virtual goods in F2P games. The proposed model can provide insights for F2P game providers to design their games and marketing strategies.

Appendices

Appendix 1: Demographic profile of players

Demographics	Frequency (n = 352)	Percentage (%)
Gender
Male	107	30.4
Female	220	62.5
Non-binary/third gender	21	6.0
Prefer not to say	4	1.1
Age
Between 18 and 20	69	19.6
Between 21 and 30	176	50.0
Between 31 and 40	75	21.3
Between 41 and 50	25	7.1
Between 51 and 60	5	1.4
Over 60	2	0.6
Education
High school	104	29.5
College	126	35.8
Bachelor	94	26.7
Master	28	8.0
Income
Less than 1000$	119	33.8
Between 1000$–1500$	51	14.5
Between 1500$–2500$	62	17.6
Between 2500$–3500$	49	13.9
Between 3500$–5000$	30	8.5
More than 5000$	41	11.6
How long are you playing the game?
0–3 hours	120	34.1
3–6 hours	119	33.8
6–9 hours	47	13.4
9–12 hours	29	8.2
More than 12 hours	37	10.5
How much, on average, are you willing to pay for virtual goods?
Less than10 $	186	52.8
11–25 $	73	20.7
26–50 $	33	9.34
51–100 $	32	9.12
101–250 $	15	4.30
More than 250 $	13	3.74

Appendix 2: Measurement items

Price adapted from (Kim et al. 2011)

PR1: The virtual goods sold in the F2P games are generally reasonably priced

PR2: The virtual goods sold in the F2P games offer value for money

PR3: The virtual goods sold in the F2P games are good products for the price

PR4: The virtual goods sold in the F2P games are considered economical in terms of price

Quality adapted from (Kim et al. 2011)

QUAL1: The virtual goods Sold in the F2P games have an acceptable standard of quality

QUAL2: The virtual goods sold in the F2P games are reliable in their performance

QUAL3: The virtual goods sold in the F2P games are good in terms of their overall excellence

QUAL: The virtual goods sold in the F2P games possess a degree of quality that is satisfactory

Achievement adapted from (Li et al. 2015)

ACH1: I can beat/surpass other players in the F2P games due to purchasing virtual goods

ACH2: I gain more power than others in the F2P games due to purchasing virtual goods

ACH3: I get a higher status/degree than other players in the F2P games due to purchasing virtual goods

Enjoyment adapted from (Li et al. 2015)

ENJ1: I feel the activity of purchasing virtual goods is interesting

ENJ2: I am happy to purchase virtual goods in the F2P games

ENJ3: I am happy to use virtual goods that I purchased in the F2P games

ENJ4: I enjoy using virtual goods that I purchased in the F2P games

Aesthetics adapted from (Kim et al. 2011)

AES1: The virtual goods sold in the F2P games are lovely

AES2: The virtual goods sold in the F2P games reflect the beauty

AES3: The virtual goods sold in the F2P games are aesthetically appealing

AES4: The virtual goods sold in the F2P games have an attractive aesthetic feature

Customization adapted from (Teng 2010)

CZ1: I have more items in the game because I purchased virtual goods

CZ2: I can modify the appearance and many goods in the F2P game because I purchased virtual goods

CZ3: I can change many things about my game following my preferences because I purchased virtual goods

Social presence adapted from (Li et al. 2015)

SOC1: I can offer more help to others using the Virtual goods I purchased in the F2P games

SOC2: I can be myself and show what kind of player/person I am by purchasing virtual goods in the F2P games

SOC3: I feel like I am a member of the F2P games community because of the virtual goods I purchased

SOC1: I feel connected to other players in the F2P games due to using virtual goods

Perceived scarcity adapted from (Chen and Sun 2014)

SC1: In my opinion, the limited virtual goods are going to be sold out soon

SC2: I think the limited virtual goods surely attract more people to buy than the available virtual items

SC3: The number of limited virtual goods is very limited

SC4: It is difficult to acquire the limited virtual goods

SC5: The limited virtual goods in the F2P game are scarce

Self-presentation adapted from (Lee et al. 2012)

SL1: I use virtual goods in the game because it helps other players to perceive me as competent

SL2: I use virtual goods in the game because it helps other players to perceive me as socially desirable

SL3: I use virtual goods in the game because it helps other players to perceive me as likable

SL4: I use virtual goods in the game because it helps other players to perceive me as friendly

SL5: I use virtual goods in the game because it helps other players to perceive me as skilled

SL6: I use virtual goods in the game because it helps me to make a good impression

SL7: I use virtual goods in the game because it helps me to tell others a little bit about myself

Perceived value adapted from (Yang et al. 2016)

SL1: Using virtual goods in the F2P games is a good deal

SL2: Compared to the effort I make, using virtual in the F2P games is beneficial to me

SL3: Compared to the time I spend, virtual goods in the F2P games are worthwhile

SL4: Overall, using virtual goods in the F2P games delivers good value

Purchase Intention adapted from (Guo and Barnes 2012)

PI1: I intend to purchase virtual goods for my characters in online games

PI2: My willingness to buy advanced virtual goods in online games is high

PI3: The likelihood that I would purchase advanced goods in online games is high

Appendix 3: ULMC test results

Construct	Item	Substantive method loading (R1)	R12	Method factor loading (R2)	R22
Price (FCT=1.286)	PR1	0.785**	0.616	− 0.047	0.002
	PR2	0.717**	0.514	0.087	0.008
	PR3	0.841**	0.707	0.077	0.006
	PR4	0.732**	0.536	0.091	0.008
Quality (FCT=1.093)	QUAL1	0.737**	0.543	− 0.003	0.000
	QUAL2	0.778**	0.605	− 0.016	0.000
	QUAL3	0.767**	0.588	0.046	0.002
	QUAL4	0.815**	0.664	− 0.029	0.001
Achievement (FCT=1.284)	ACH1	0.880**	0.774	− 0.033	0.001
	ACH2	0.911**	0.830	0.002	0.000
	ACH3	0.767**	0.588	0.034	0.001
Enjoyment (FCT=1.075)	ENJ1	0.603**	0.364	0.075**	0.006
	ENJ2	0.524**	0.275	0.259*	0.067
	ENJ3	0.881**	0.776	− 0.131*	0.017
	ENJ4	0.878**	0.771	− 0.171**	0.029
Aesthetics (FCT=1.684)	AES1	0.724**	0.524	0.016	0.000
	AES2	0.681**	0.464	0.130*	0.017
	AES3	0.885**	0.783	− 0.132*	0.017
	AES4	0.815**	0.664	− 0.013	0.000
Customization (FCT=1.356)	CZ1	0.727**	0.529	0.078	0.006
	CZ2	0.866**	0.750	− 0.047	0.002
	CZ3	0.818**	0.669	− 0.027	0.001
Social presence (FCT=1.036)	SOC1	0.509**	0.259	0.060	0.004
	SOC2	0.723**	0.523	0.054	0.003
	SOC3	0.854**	0.729	− 0.049	0.002
	SOC4	0.848**	0.719	− 0.042	0.002
Perceived scarcity (FCT=1.101)	SC1	0.668**	0.446	0.178	0.032
	SC2	0.569**	0.324	− 0.064	0.004
	SC3	0.785**	0.616	− 0.088	0.008
	SC4	0.793**	0.629	− 0.019	0.000
	SC5	0.828**	0.686	0.034*	0.001
Self-presentation (FCT=1.219)	SL1	0.777**	0.604	− 0.025	0.001
	SL2	0.750**	0.563	0.012	0.000
	SL3	0.821**	0.674	− 0.018	0.000
	SL4	0.670**	0.449	0.100	0.010
	SL5	0.712**	0.507	− 0.017	0.000
	SL6	0.852**	0.726	− 0.053	0.003
	SL7	0.665**	0.442	0.006	0.000
Perceived value (FCT=1.516)	PV1	0.766**	0.587	− 0.025	0.001
	PV2	0.791**	0.626	0.008	0.000
	PV3	0.817**	0.667	− 0.079	0.006
	PV4	0.850**	0.723	0.038*	0.001
Purchase Intention (FCT=1.667)	PI1	0.594**	0.353	− 0.002	0.000
	PI2	0.935**	0.874	− 0.018	0.000
	PI3	0.955**	0.912	− 0.122	0.015
Average			0.604		0.006

1. **p < 0.01; *p < 0.05

Appendix 4: Discriminant validity—Fornell–Larcker criterion

Construct	PR	QUAL	ACH	ENJ	AES	CZ	SOC	SC	SL	PV	PI
PR	0.588
QUAL	0.042	0.599
ACH	0.047	0.107	0.730
ENJ	0.112	0.143	0.057	0.520
AES	0.097	0.179	0.116	0.252	0.602
CZ	0.056	0.084	0.097	0.159	0.201	0.646
SOC	0.070	0.289	0.102	0.123	0.127	0.141	0.555
SC	0.092	0.045	0.033	0.046	0.050	0.029	0.098	0.532
SL	0.057	0.200	0.146	0.155	0.208	0.170	0.246	0.055	0.564
PV	0.128	0.359	0.210	0.299	0.312	0.226	0.265	0.037	0.374	0.650
PI	0.176	0.297	0.117	0.265	0.304	0.179	0.338	0.138	0.319	0.399	0.701

1. Values in Italic represents Square root of AVE. Square correlations; AVE in the diagonal
2. PR price, QUAL quality, ACH achievement, ENJ enjoyment, AES aesthetics, CZ customization, SOC social presence, SC perceived scarcity, SL self-presentation, PV perceived value, PI purchase intention