The impact of brand communication on brand equity through Facebook

Journal of Research in Interactive Marketing, 2015

Bruno Schivinski, Dariusz Dabrowski

Journal of Research in Interactive Marketing, volume 9, issue 1, pages 31–53, March 2015

  • 302 Facebook users
  • 60 brands, 3 industries
  • Structural equation modelling
  • Poland
Download the accepted manuscript (PDF)Published versionUse the measures

Free full text: the authors’ accepted manuscript, identical in content to the published article. Please cite the published version.

Keywords
  • social media marketing
  • Facebook
  • social networking sites
  • structural equation modelling
  • marketing communication
  • brand equity
  • firm-created social media communication
  • user-generated content (UGC)
  • brand-related user-generated content
  • social media brand communication
  • brand fan pages
  • electronic word-of-mouth (eWOM)
  • consumer-based brand equity (CBBE)
  • brand awareness
  • brand associations
  • brand loyalty
  • perceived quality
  • Aaker brand equity model
  • hierarchy of effects
  • consumer learning hierarchy
  • advertising effects on brand equity
  • online survey
  • confirmatory factor analysis
  • multi-group analysis
  • critical ratio difference test (CRDIFF)
  • AMOS
  • Poland
  • non-alcoholic beverages
  • clothing brands
  • mobile network providers
  • industry differences
  • social media strategy
  • brand management
  • digital marketing
  • content marketing

In a survey of 302 Polish Facebook users who had “Liked” 60 brands in three industries, content that other users post about a brand raised brand awareness and associations, brand loyalty and perceived quality, while content the brand posts itself raised only awareness and associations. The effects differed between non-alcoholic beverages, clothing and mobile network providers.

Key findings

  1. Brand-posted content builds awareness only. Firm-created social media communication had a positive effect on brand awareness/associations (β = 0.14, p = 0.02) but no significant effect on brand loyalty (β = −0.08, p = 0.15) or perceived quality (β = −0.03, p = 0.61) among 302 Facebook users (Schivinski & Dabrowski, 2015).
  2. User-generated content strengthens all three dimensions. User-generated social media communication had positive effects on brand awareness/associations (β = 0.12, p = 0.05), brand loyalty (β = 0.24, p = 0.001) and perceived quality (β = 0.26, p = 0.001) (Schivinski & Dabrowski, 2015).
  3. Awareness leads to loyalty and perceived quality. Brand awareness/associations had positive effects on brand loyalty (β = 0.13, p = 0.03) and perceived quality (β = 0.22, p = 0.001), consistent with a hierarchy-of-effects view of brand equity (Schivinski & Dabrowski, 2015).
  4. Effects differ by industry. The effect of firm-created communication on brand awareness/associations was largest for non-alcoholic beverages (β = 0.31, p = 0.003) and not significant for clothing, while user-generated communication affected perceived quality strongly for mobile network providers (β = 0.51, p = 0.001), significantly more than in the other two industries (Schivinski & Dabrowski, 2015).
  5. Valid and reliable measures. The five-factor measurement model fitted well (χ²/df = 1.54, CFI = 0.98, TLI = 0.98, RMSEA = 0.04, SRMR = 0.03), with composite reliabilities of 0.85 to 0.95 and average variance extracted of 0.59 to 0.86 (Schivinski & Dabrowski, 2015).

Use the measures

Social media brand communication and brand equity measures

The full set of measures used in the article: firm-created and user-generated social media brand communication (four items each), brand awareness/associations (four items), brand loyalty (three items) and perceived quality (three items), 18 items in total. Respondents name a brand they follow on Facebook and answer on a seven-point agreement scale. Free to use in research and practice; please cite the article.

Items
18 in five constructs: firm-created (4) and user-generated (4) social media communication, brand awareness/associations (4), brand loyalty (3), perceived quality (3)
Response format
7-point Likert scale from 1 = strongly disagree to 7 = strongly agree
Scoring
Mean of the items in each construct (range 1–7)
Reliability
Cronbach’s α = 0.84 to 0.95; composite reliability = 0.85 to 0.95; AVE = 0.59 to 0.86
Validated with
302 Facebook users in Poland, 60 brands in three industries

Ready-to-use files

The questionnaire

Introduction suggested wording

This survey is about a brand you follow on Facebook. Brands post their own content on their Facebook pages, such as photos, videos and news from the company. Other Facebook users also post content about brands, such as comments, photos and reviews. There are no right or wrong answers.

Part 1. Your brand suggested wording

Please type the name of a brand that you have “Liked” on Facebook.

You will answer the following questions about this brand. Wherever you see [brand], think of the brand you named.

Part 2. Firm-created social media communication Brand communication on social media created and managed by the company.

Think about [brand]. Please indicate how much you agree or disagree with each statement. suggested wording

Item 1Strongly disagree 2 3 4 5 6 7Strongly agree
FC1I am satisfied with the company’s social media communications for [brand].
FC2The level of the company’s social media communications for [brand] meets my expectations.
FC3The company’s social media communications for [brand] are very attractive.
FC4This company’s social media communications for [brand] perform well, when compared with the social media communications of other companies.

Part 3. User-generated social media communication Content about the brand created and posted by other social media users, outside the company’s control.

Think about [brand]. Please indicate how much you agree or disagree with each statement. suggested wording

Item 1Strongly disagree 2 3 4 5 6 7Strongly agree
UG1I am satisfied with the content generated on social media sites by other users about [brand].
UG2The level of the content generated on social media sites by other users about [brand] meets my expectations.
UG3The content generated by other users about [brand] is very attractive.
UG4The content generated on social media sites by other users about [brand] performs well, when compared with other brands.

Part 4. Brand awareness/associations How easily consumers recognise and recall the brand and the associations it brings to mind.

Think about [brand]. Please indicate how much you agree or disagree with each statement. suggested wording

Item 1Strongly disagree 2 3 4 5 6 7Strongly agree
BAS1I easily recognize [brand].
BAS2Several characteristics of [brand] instantly come to my mind.
BAS3I can quickly recall the symbol or logo of [brand].
BAS4I can recognize [brand] among other competing brands.

Part 5. Brand loyalty Commitment to keep buying from the brand.

Think about [brand]. Please indicate how much you agree or disagree with each statement. suggested wording

Item 1Strongly disagree 2 3 4 5 6 7Strongly agree
BL1RThe prospect of lower prices would make me switch to another company.
BL2RIf it were possible to do so without problems, I would choose another company.
BL3I intend to remain the company’s customer.

Part 6. Perceived quality The consumer’s judgement of the brand’s overall quality.

Think about [brand]. Please indicate how much you agree or disagree with each statement. suggested wording

Item 1Strongly disagree 2 3 4 5 6 7Strongly agree
PQ1Most of the products of [brand] are of great quality.
PQ2The likelihood that [brand] is reliable is very high.
PQ3Products of [brand] are worth their price.

Item wording, response format and sources are as published in the article’s Table AI (BAS4 printed “X” for the brand name; here it reads [brand]). The introduction, brand question and instruction are suggested wording based on the procedure the article describes.

How to administer

  • Ask respondents to name a brand they have “Liked” on Facebook, and insert that name into every item (for example with piped text in Qualtrics).
  • Give a short example of each type of content (posts by the brand, posts by other users) so that respondents can tell firm-created and user-generated communication apart, as in the original study.
  • The original study also screened respondents on how often they receive brand content and read what other users post about the brand; include similar screening questions if exposure matters for your study.
  • The items work for other social networks: replace “social media sites” with the platform you study, keeping the rest of the wording.
  • For use in another language, translate and back-translate the items (the original was administered in Polish) and check measurement invariance before comparing groups.

How to score

  • Code responses 1 (strongly disagree) to 7 (strongly agree).
  • Compute one score per construct as the mean of its items: firm-created = mean(FC1–FC4); user-generated = mean(UG1–UG4); brand awareness/associations = mean(BAS1–BAS4); brand loyalty = mean(BL1–BL3); perceived quality = mean(PQ1–PQ3).
  • BL1 and BL2 are reverse-scored (as in the original study): recode them as 8 − x before computing the brand loyalty score, so that higher scores mean higher loyalty. They are marked R in the questionnaire above.
  • In the original data, item means ranged from 4.42 to 5.22 for the communication items and from 5.58 to 6.78 for the brand equity items (Table AI).
  • For latent-variable models, specify five correlated factors, then the structural model in Figure 2.

At a glance

Design
Online survey posted on Facebook brand fan pages; structural equation modelling and multi-group comparison across industries
Participants
302 Polish Facebook users (56.7% women; most aged 15–35) who had “Liked” one of 60 brands
Industries
Non-alcoholic beverages (n = 99), clothing (n = 99) and mobile network providers (n = 104)
Measures
7-point Likert scales: firm-created and user-generated social media communication (4 items each), brand awareness/associations (4), brand loyalty (3), perceived quality (3)
Analysis
Exploratory and confirmatory factor analysis, structural equation modelling and critical ratio difference (CRDIFF) tests in AMOS 21.0

Figure 1. Conceptual framework

H2aH2bH1aH1bH3aH3bH4H5Firm-createdsocial mediacommunicationUser-generatedsocial mediacommunicationBrand loyaltyBrand awareness/associationsPerceived quality
Firm-created and user-generated social media brand communication are expected to influence brand awareness/associations (H1a, H1b), brand loyalty (H2a, H2b) and perceived quality (H3a, H3b); brand awareness/associations is expected to influence brand loyalty (H4) and perceived quality (H5).

Figure 2. Parameter estimates for the final structural model

H2a: n.s.H2b: 0.24***H1a: 0.14**H1b: 0.12*H3a: n.s.H3b: 0.26***H4: 0.13**H5: 0.22***Firm-createdsocial mediacommunicationUser-generatedsocial mediacommunicationBrand loyaltyBrand awareness/associationsPerceived quality
Standardised coefficients; dashed grey paths are not significant (n.s.). * p < 0.05; ** p < 0.01; *** p < 0.001. Model fit: Cmin/df = 2.53, CFI = 0.95, TLI = 0.94, RMSEA = 0.07, SRMR = 0.07.

Tables

Table I. Correlation matrix and indicators of reliability and validity

Constructs and measurements α CR AVE MSV ASV UG BAW/BAS FC PQ BL
UG 0.946 0.920 0.744 0.325 0.123 0.863
BAW/BAS 0.836 0.849 0.589 0.067 0.043 0.198 0.767
FC 0.944 0.944 0.809 0.325 0.099 0.570 0.206 0.899
PQ 0.891 0.897 0.744 0.148 0.080 0.285 0.259 0.156 0.863
BL 0.924 0.947 0.856 0.148 0.056 0.219 0.155 0.072 0.385 0.925

Notes: The square root of the average variance extracted (AVE) values are marked in italics; FC = firm-created social media communication; UG = user-generated social media communication; BAW/BAS = brand awareness/associations; BL = brand loyalty; PQ = perceived quality

Table II. Standardized structural coefficients of the model

Hypothesis β t-value p-value Acceptance or rejection
H1a. Firm-created social media → Brand awareness/associations 0.14 2.30 0.02 Accepted
H1b. User-generated social media → Brand awareness/associations 0.12 1.93 0.05 Accepted
H2a. Firm-created social media → Brand loyalty −0.08 −1.40 0.15 Rejected
H2b. User-generated social media → Brand loyalty 0.24 3.94 0.001 Accepted
H3a. Firm-created social media → Perceived quality −0.03 −0.50 0.61 Rejected
H3b. User-generated social media → Perceived quality 0.26 4.19 0.001 Accepted
H4. Brand awareness/associations → Brand loyalty 0.13 2.11 0.03 Accepted
H5. Brand awareness/associations → Perceived quality 0.22 3.45 0.001 Accepted

Notes: Cmin/df = 2.53; CFI = 0.95; TLI = 0.94; RMSEA = 0.07 (90 % C.I. 0.06, 0.08); SRMR = 0.07

Table III. Results of the industry comparison
Path Beverages unstd. β Beverages std. β p Clothing unstd. β Clothing std. β p Mobile unstd. β Mobile std. β p B × C z B × M z C × M z
FC → BAW/BAS 0.126 0.312 0.003 0.043 0.085 0.435 0.103 0.235 0.026 −1.192 −0.377 0.827
UG → BAW/BAS −0.009 −0.021 0.843 0.116 0.190 0.090 0.000 −0.001 0.996 1.524 0.146 −1.472
UG → BL 0.270 0.262 0.011 0.171 0.160 0.142 0.176 0.186 0.079 −0.626 −0.644 0.031
UG → PQ 0.093 0.137 0.162 0.064 0.077 0.488 0.465 0.519 0.001 −0.247 3.259*** 3.045***
BAW/BAS → PQ 0.737 0.470 0.001 0.277 0.202 0.078 0.217 0.091 0.338 −1.944* −1.809* −0.217
BAW/BAS → BL 0.360 0.151 0.152 0.350 0.199 0.070 0.265 0.105 0.319 −0.031 −0.259 −0.259

Notes: FC = firm-created social media brand communication; UG = user-generated social media brand communication; BAW/BAS = brand awareness/associations; BL = brand loyalty; PQ = perceived quality; B = non-alcoholic beverages industry; C = clothing industry; M = mobile network providers; Cmin/df = 1.75; CFI = 0.93; TLI = 0.92; RMSEA = 0.05 90 per cent C.I. 0.04 0.05; SRMR = 0.07; * < 0.05; *** < 0.001

Table AI. List of constructs and measurements used
Constructs and measurements Standardized factor loading t-value Mean SD Authors
Firm-created social media communication Mägi (2003); Tsiros et al. (2004); Bruhn et al. (2012)
[FC1] I am satisfied with the company’s social media communications for [brand] 0.90 22.31 5.22 1.29
[FC2] The level of the company’s social media communications for [brand] meets my expectations 0.91 23.13 5.18 1.32
[FC3] The company’s social media communications for [brand] are very attractive* 0.91 22.96 5.02 1.31
[FC4] This company’s social media communications for [brand] perform well, when compared with the social media communications of other companies 0.88 a 5.07 1.23
User-generated social media communication Mägi (2003); Tsiros et al. (2004); Bruhn et al. (2012)
[UG1] I am satisfied with the content generated on social media sites by other users about [brand] 0.89 17.41 4.73 1.29
[UG2] The level of the content generated on social media sites by other users about [brand] meets my expectations 0.91 17.83 4.71 1.24
[UG3] The content generated by other users about [brand] is very attractive* 0.81 a 4.42 1.34
[UG4] The content generated on social media sites by other users about [brand] performs well, when compared with other brands 0.85 18.68 4.70 1.22
Brand awareness/association Yoo et al. (2000); Villarejo-Ramos and Sánchez-Franco (2005)
[BAS1] I easily recognize [brand] 0.75 14.66 6.78 0.47
[BAS2] Several characteristics of [brand] instantly come to my mind 0.63 11.27 6.35 0.64
[BAS3] I can quickly recall the symbol or logo of [brand] 0.74 14.35 6.63 0.55
[BAS4] I can recognize X among other competing brands 0.93 a 6.65 0.54
Brand loyalty Walsh et al. (2009)
[BL1] The prospect of lower prices would make me switch to another company 0.93 28.01 5.72 1.15
[BL2] If it were possible to do so without problems, I would choose another company 0.92 27.89 5.60 1.10
[BL3] I intend to remain the company’s customer 0.92 a 5.58 1.15
Perceived quality Yoo et al. (2000)
[PQ1] Most of the products of [brand] are of great quality 0.86 16.76 5.84 0.99
[PQ2] The likelihood that [brand] is reliable is very high 0.91 17.39 5.67 0.97
[PQ3] Products of [brand] are worth their price 0.81 a 5.63 1.11

Notes: *New item from the authors; apath constrained to 1 for model specification

The question

Brands use Facebook to communicate with consumers, and consumers post their own content about brands on the same platform. Research had studied eWOM, online reviews and brand communities, but no study had examined how firm-created and user-generated social media communication each affect the dimensions of consumer-based brand equity: brand awareness/associations, perceived quality and brand loyalty. Nor had any study compared these effects across industries.

Implications

Practical implications

Posting brand content builds awareness and associations, but on its own it does not make consumers more loyal or raise their view of quality; content from other users does. Managers should therefore keep publishing their own content while encouraging consumers to create and share brand content, for example through exclusive campaigns, contests and offers that require users to engage with the fan page. Strategies should be tailored by industry: in the study, firm content worked best for non-alcoholic beverages, while user content shaped perceived quality most for mobile network providers.

Theoretical implications

The study separates two sources of social media brand communication and shows that they affect consumer-based brand equity differently: user-generated communication influences all three dimensions, firm-created communication only awareness/associations. It supports the hierarchy-of-effects view in which awareness/associations precede loyalty and perceived quality, and shows that industry moderates the effects of social media communication on brand equity.

Limitations and future research

  • Only Facebook was studied; future research should compare the leading social networks, as brand communication differs across channels.
  • Only three industries were examined; a broader range would show how consumers perceive brands from different industries on social media.
  • Each construct was measured with a small number of items; replications should consider adding items.
  • The sample was Polish and mostly young, which limits generalisation; studies in other countries should account for social, economic and cultural differences.
  • Future research could use Keller’s brand equity framework, control for differences between brands, and link social media communication to company financial performance.

Questions this paper answers

Does user-generated content on social media affect brand equity?

Yes. In a study of 302 Facebook users, user-generated social media communication had positive effects on brand awareness/associations (β = 0.12), brand loyalty (β = 0.24) and perceived quality (β = 0.26) (Schivinski & Dabrowski, 2015).

Does a brand’s own Facebook content make consumers more loyal?

Not directly. Firm-created social media communication raised brand awareness/associations (β = 0.14) but had no significant effect on brand loyalty or perceived quality; it can affect them indirectly, because awareness influences loyalty and perceived quality (Schivinski & Dabrowski, 2015).

Do the effects of social media communication differ by industry?

Yes. The effect of firm-created communication on awareness/associations was largest for non-alcoholic beverages (β = 0.31), and user-generated communication shaped perceived quality much more for mobile network providers (β = 0.51) than for the other industries (Schivinski & Dabrowski, 2015).

How can I measure firm-created and user-generated social media communication?

With two four-item scales adapted from Mägi (2003), Tsiros et al. (2004) and Bruhn et al. (2012), each with one new item, answered on a 7-point agreement scale; both had composite reliability above 0.90 (Schivinski & Dabrowski, 2015).

Key concepts

Firm-created social media communication
Brand communication on social media that is created and controlled by the company, for example posts on the brand’s Facebook page.
User-generated social media communication
Brand-related content made publicly available online by users, reflecting some creative effort and created outside professional routines and practices (the OECD definition of user-generated content used in the article).
Consumer-based brand equity (CBBE)
The value a brand holds in consumers’ minds; the article measures three of Aaker’s dimensions: brand awareness/associations, perceived quality and brand loyalty.
Critical ratio difference (CRDIFF) test
A pairwise test of whether a path coefficient differs between groups in a multi-group structural model, reported with z-values.

Open science and materials

How to cite

Schivinski, B., & Dabrowski, D. (2015). The impact of brand communication on brand equity through Facebook. Journal of Research in Interactive Marketing, 9(1), 31–53. https://doi.org/10.1108/JRIM-02-2014-0007

Related publications

All publications

Buy me a coffee

Loading…

Open on Ko-fi