Влияние лидеров мнений на коммуникацию бренда в сфере красоты на примере копании M.A.C Cosmetics в России

14. Choi, B., Lee, I. Trust in open versus closed social media: The relative influence of user- and marketer-generated content in social network services on customer trust. // Telematics and Informatics. 2016. №34 (5). С. 550-559.

15. Farquhar, P. H. MANAGING BRAND EQUITY // Marketing Research. 1989. №1 (3). С. 24-33.

16. Flight, M. Companies Tap to Consumer Passion // Business 2.0. 2005.

17. Fournier, S., Avery J. The uninvited brand // Business Horizons. 2011. №54 (3). С. 193-207.

18. Freberg, K., Graham, K., McGaughey, K., Freberg, L. A. Who are the social media influencers? A study of public perceptions of personality. // Public Relations Review. 2011. №37 (1). С. 90-92.

19. Fulgoni, G. M., Lipsman, A. Digital Word of Mouth And Its Offline Amplification // Journal of Advertising Research. 2015. №55 (1). С. 18-21.

20. Gladwell, M. The Tipping Point : How Little Things Can Make a Big Difference. Boston: Back Bay Books, 2002.

21. Global influencer marketing: insight from beauty. // Traackr. 2017

22. Goh, K., Heng, C., Lin, Z. Social Media Brand Community and Consumer Behavior: Quantifying the Relative Impact of User- and Marketer-Generated Content // Information Systems Research. 2013. №24 (1). С. 88-107.

23. Higie, R. A., Feick, L. F., Price, L. L. Types and Amount of Word-of-Mouth Communications About Retailers // Journal of Retailing. 1987. №63 (3).

24. Hutter, K., Hautz, J., Dennhardt, S., Fьller, J. The impact of user interactions in social media on brand awareness and purchase intention: the case of MINI on Facebook // Journal of Product & Brand Management. 2013. №22 (5). С. 342-351.

25. Jin, S., Phua, J. Following Celebrities' Tweets About Brands: The Impact of Twitter-Based Electronic Word-of-Mouth on Consumers' Source Credibility Perception, Buying Intention, and Social Identification With Celebrities. // Journal of Advertising. 2014. №43 (2). С. 181-195.

26. Karakaya, F., Barnes, N.G. Impact of online reviews of customer care experience on brand or company selection // Journal of Consumer Marketing. 2010. №27 (5). С. 447-457.

27. Keller, K. L. Conceptualizing, Measuring, Managing Customer-Based Brand Equity // Journal of Marketing. 1993. №57 (1).

28. Keller, E., Berry, J. The Influentials: One American in Ten Tells the Other Nine How to Vote, Where to Eat, and What to Buy. Ney York: Free Press, 2003.

29. Kelly, L., Kerr, G., Drennan, J. Avoidance of advertising in social networking sites: The teenage perspective // Journal of Interactive Advertising. 2010. №10. С. 16-27.

30. Khamis, S., Ang, L., Welling, R. Self-Branding, `micro-Celebrity' and the Rise of Social Media Influencers // Celebrity Studies. 2017. №8 (2). С. 191-208.

31. Kirmani, A.,Wright, P. Money talks: perceived advertising expense and expected product quality // Journal of Consumer Research. 1989. №16 (3). С. 344-353.

32. Kotler, P., Keller, K. L. Marketing Management. 15 изд. Harlow: Pearson, 2016.

33. Lazarsfeld, P.F., Berelson, B., Gaudet, H. The people's choice: How the voter makes up his mind in a presidential campaign. New York: Columbia University Press, 1944.

34. Leigert, L. Marketing communication in social media // Marketing Research. 2013. №21 (3). С. 12-19.

35. Mangold, W. G., D. J. Faulds. Social Media: The New Hybrid Element of the Promotion Mix // Business Horizons. 2009. №52 (4). С. 357- 365.

36. Mдgi, A.W. Share of wallet in retailing: the effects of customer satisfaction, loyalty cards and shopper characteristics // Journal of Retailing. 2003. №79 (2). С. 97-106.

37. Muсiz, A.M., Schau, H.J Vigilante marketing and consumer-created communications // Journal of Advertising. 2007. №36 (3). С. 35-50.

38. Oliver, R. Satisfaction: A Behavioral Perspective on the Consumer. New York: McGraw-Hill, 1997.

39. Palmatier, R.W., Scheer, L.K., Stennkamp, J.B.E.M. Customer loyalty to whom? Managing the benefits and risks of salesperson-owned loyalty // Journal of Marketing Research. 2007. №44 (2). С. 185-199.

40. Quester, P., Lim, A.L. Product involvement/brand loyalty: is there a link? // Journal of Product and Brand Management. 2003. №12 (1).

41. Rao, A.R., Monroe, K.B. The effect of price, brand name, and store name on buyers' perceptions of product quality: an integrative review // Journal of Marketing Research. 1989. №36 (2). С. 351-358.

42. Tsiros, M., Mittal, V., Ross, W.T. The role of attributions in customer satisfaction: a reexamination // Journal of Consumer Research. 2004. №31 (2). С. 476-483.

43. Villarejo-Ramos, A.F., Sбnchez-Franco, M.J. The impact of marketing communication and price promotion on brand equity // Journal of Brand Management. 2005. №12 (6). С. 431-444.

44. Walsh, G., Mitchell, V.-W., Jackson, P.R., Beatty, S.E. Examining the antecedents and consequences of corporate reputation: a customer perspective // British Journal of Management. 2009. №20 (2). С. 187-203.

45. Yoo, B., Donthu, N., Lee, S. An examination of selected marketing mix elements and brand equity // Journal of the Academy of Marketing Science. 2000. №28 (2). С. 195-211.

Электронные ресурсы:

46. 2017 Adblock Report // pagefair.com URL: https://pagefair.com/blog/2017/adblockreport/ (дата обращения: 28.03.2018).

47. Исследование аудитории Instagram: сервисом пользуется каждый десятый в России, большинство -- женщины // forbes.ru URL: http://www.forbes.ru/tehnologii/343331-issledovanie-auditorii-instagram-servisom-polzuetsya-kazhdyy-desyatyy-v-rossii (дата обращения: 08.03.2018).

48. Опубликовано исследование о важности сегментов аудитории для брендов // prexplore.ru URL: https://prexplore.ru/news/research-about-importance-audience-for-brands/ (дата обращения: 28.03.2018).

49. Распределение упоминаний продуктов Estee Lauder и конкурентов январь-декабрь 2017 // Mediascope. 2018.

50. Based on Trust, Engagement Rates, and Influence, True Digital Creators Offer Greatest Impact for Brand Partnerships // shareablee.com URL: https://www.shareablee.com/blog/2018/03/28/fullscreen-shareablee-report-online-influence-isnt-what-it-seems/ (дата обращения: 28.032018).

51. For Influencers, Instagram Is the Clear-Cut Favorite // emarketer.com URL: https://www.emarketer.com/content/for-influencers-instagram-is-close-to-the-only-platform-that-matters (дата обращения: 06.03.2018).

52. Global social media research summary 2018 // smartinsight.com URL: https://www.smartinsights.com/social-media-marketing/social-media-strategy/new-global-social-media-research/ (дата обращения: 28.03.2018).

53. How Influencer Content Has Put Fashion And Beauty Brands Ahead On Instagram // socialbakers.com URL: https://www.socialbakers.com/blog/2757-how-influencer-content-has-put-fashion-and-beauty-brands-ahead-on-instagram (дата обращения: 02.03.2018).

54. Influence by the Numbers // fullscreenmedia.co URL: https://fullscreenmedia.co/2018/03/27/influence-numbers-lowdown-whos-really-influential-online/ (дата обращения: 28.03.2018).

55. Influencer Content Is Beautiful // tapinfluence.com URL: https://www.tapinfluence.com/beauty/ (дата обращения: 06.03.2018).

56. Influencer-generated content: your guide to using it right // izea.com URL: https://izea.com/2017/11/14/influencer-generated-content-guide-using-right/ (дата обращения: 12.03.2018).

57. INSPIRE ACTION. // instagram.com URL: https://business.instagram.com/ (дата обращения: 06.03.2018).

58. Instagram Engagement Rate Data: Average Seconds on Site // yotpo.com URL: https://www.yotpo.com/blog/instagram-engagement-rate/ (дата обращения: 08.03.2018).

59. Number of Companies Using Instagram To Nearly Double Next Year // emarketer.com URL: https://www.emarketer.com/Article/Number-of-Companies-Using-Instagram-Nearly-Double-Next-Year/1013122 (дата обращения: 06.03.2018).

60. SOCIAL INFLUENCE: MARKETING'S NEW FRONTIER // crowdtap.com URL: http://corp.crowdtap.com/socialinfluence (дата обращения: 01.02.2018).

61. The State of Influencer Marketing 2017 // Linqua. 2017.

62. Twitter Says Users Now Trust Influencers Nearly as Much as Their Friends // adweek.com URL: http://www.adweek.com/digital/twitter-says-users-now-trust-influencers-nearly-much-their-friends-171367/ (дата обращения: 01.02.2018).

63. WHY INFLUENCER GENERATED CONTENT IS THE NEW UGC // revfluence.com URL: http://blog.revfluence.com/blog/2017/8/10/why-influencer-generated-content-is-the-new-ugc (дата обращения: 13.04.2018).



Приложение 1. Скриншоты онлайн-анкеты



Приложение 2. Скрипт конфирматорного факторного анализа

Установление рабочей директории, загрузка базовых пакетов и загрузка базы данных.

title: "Diploma Elizaveta Shved" author: "Elizaveta Shved" date: "6 05 2018" output: pdf_document: default html_document: default word_document: default header-includes: - - - fontsize: 12pt ---

#Установим рабочую директорию, а также загрузим базовые пакеты для дальнейшего анализа

getwd()

## [1] "/Users/eshvedova/Desktop/Документы"

library(rmarkdown)

library(lavaan)

## This is lavaan 0.5-23.1097

## lavaan is BETA software! Please report any bugs.

library(sjmisc)

library(nFactors)

## Loading required package: MASS

## Loading required package: psych

##

## Attaching package: 'psych'

## The following object is masked from 'package:lavaan':

##

## cor2cov

## Loading required package: boot

##

## Attaching package: 'boot'

## The following object is masked from 'package:psych':

##

## logit

## Loading required package: lattice

##

## Attaching package: 'lattice'

## The following object is masked from 'package:boot':

##

## melanoma

##

## Attaching package: 'nFactors'

## The following object is masked from 'package:lattice':

##

## parallel

library(ggplot2)

##

## Attaching package: 'ggplot2'

## The following objects are masked from 'package:psych':

##

## %+%, alpha

library(sjPlot)

## #refugeeswelcome

library(GPArotation)

library(semPlot)

library(haven)

library(tidyr)

##

## Attaching package: 'tidyr'

## The following object is masked from 'package:sjmisc':

##

## replace_na

library(dplyr)

##

## Attaching package: 'dplyr'

## The following object is masked from 'package:MASS':

##

## select

## The following objects are masked from 'package:stats':

##

## filter, lag

## The following objects are masked from 'package:base':

##

## intersect, setdiff, setequal, union

library(magrittr)

##

## Attaching package: 'magrittr'

## The following object is masked from 'package:tidyr':

##

## extract

library(sjmisc)

library(ggplot2)

library(kableExtra)

## Warning: package 'kableExtra' was built under R version 3.4.4

# Откроем базу данных

Data <-read.csv('massive.csv', sep = ";", header=TRUE, na.strings= "99")

# Выберем переменные из массива

Mac.values <- Data %>%

dplyr::select("Var1",

"Var2",

"Var3",

"Var4",

"Var5",

"Var6",

"Var7",

"Var7",

"Var8",

"Var9",

"Var10",

"Var11",

"Var12",

"Var13",

"Var14",

"Var15",

"Var16",

"Var17",

"Var18") %>%

na.omit

data <- Data %>%

dplyr::select("Age","Gender","Income", "beauty")%>%

na.omit

Построение матрицы коллеряций между суждениями

library(corrplot)

cor<-cbind(Mac.values$Var1, Mac.values$Var2,Mac.values$Var3, Mac.values$Var4,Mac.values$Var5, Mac.values$Var6,Mac.values$Var7,Mac.values$Var8, Mac.values$Var9, Mac.values$Var10,Mac.values$Var11, Mac.values$Var12, Mac.values$Var13, Mac.values$Var14, Mac.values$Var15, Mac.values$Var16, Mac.values$Var17, Mac.values$Var18

)

r<-cor(na.omit(cor))

corrplot(r)

Разведывательный факторный анализ.

Построение графика каменистой осыпи для определения количества факторов.

plot(nScree(cov(Mac.values)))

Построение пятифакторной факторной модели

formulaEFA <- as.formula("~Var1+Var2+Var3+ Var4 + Var5 + Var6+ Var7+ Var8+Var9+Var10+Var11+Var12+Var13+Var14+Var15+Var16+Var17+Var18")

EFA5 <- factanal(formulaEFA, factors=5, data=Mac.values,

na.action=na.exclude)

EFA5

##

## Call:

## factanal(x = formulaEFA, factors = 5, data = Mac.values, na.action = na.exclude)

##

## Uniquenesses:

## Var1 Var2 Var3 Var4 Var5 Var6 Var7 Var8 Var9 Var10 Var11 Var12

## 0.982 0.305 0.377 0.298 0.415 0.419 0.576 0.482 0.230 0.393 0.594 0.584

## Var13 Var14 Var15 Var16 Var17 Var18

## 0.241 0.928 0.576 0.435 0.350 0.279

##

## Loadings:

## Factor1 Factor2 Factor3 Factor4 Factor5

## Var1 -0.118

## Var2 0.177 0.778 0.124 0.198

## Var3 0.134 0.751 0.194

## Var4 0.134 0.783 0.240

## Var5 0.173 0.372 0.644

## Var6 0.117 0.209 0.118 0.705 0.111

## Var7 0.167 0.179 0.584 0.115

## Var8 0.196 0.332 0.555 0.231

## Var9 0.213 0.154 0.139 0.824

## Var10 0.181 0.740 0.110 0.111

## Var11 0.247 0.105 0.565

## Var12 0.101 0.625

## Var13 0.859

## Var14 -0.260

## Var15 0.562 0.115 0.110 0.281

## Var16 0.669 0.165 0.250 0.147

## Var17 0.758 0.126 0.156 0.188

## Var18 0.790 0.125 0.199 0.199

##

## Factor1 Factor2 Factor3 Factor4 Factor5

## SS loadings 2.299 2.224 2.196 1.413 1.403

## Proportion Var 0.128 0.124 0.122 0.078 0.078

## Cumulative Var 0.128 0.251 0.373 0.452 0.530

##

## Test of the hypothesis that 5 factors are sufficient.

## The chi square statistic is 294.03 on 73 degrees of freedom.

## The p-value is 2.64e-28

Построение шестифакторной модели

formulaEFA <- as.formula("~Var1+Var2+Var3+ Var4 + Var5 + Var6+ Var7+ Var8+Var9+Var10+Var11+Var12+Var13+Var14+Var15+Var16+Var17+Var18")

EFA6 <- factanal(formulaEFA, factors=6, data=Mac.values,

na.action=na.exclude)

EFA6

##

## Call:

## factanal(x = formulaEFA, factors = 6, data = Mac.values, na.action = na.exclude)

##

## Uniquenesses:

## Var1 Var2 Var3 Var4 Var5 Var6 Var7 Var8 Var9 Var10 Var11 Var12

## 0.980 0.304 0.371 0.300 0.411 0.417 0.581 0.476 0.228 0.391 0.585 0.582

## Var13 Var14 Var15 Var16 Var17 Var18

## 0.246 0.917 0.546 0.005 0.219 0.323

##

## Loadings:

## Factor1 Factor2 Factor3 Factor4 Factor5 Factor6

## Var1 -0.103

## Var2 0.127 0.779 0.106 0.207 0.114

## Var3 0.756 0.114 0.190

## Var4 0.782 0.246

## Var5 0.354 0.661 0.127

## Var6 0.121 0.207 0.712

## Var7 0.107 0.181 0.149 0.582 0.106

## Var8 0.327 0.120 0.576 0.219 0.120

## Var9 0.130 0.178 0.826 0.169

## Var10 0.748 0.125 0.114 0.106

## Var11 0.577 0.112 0.215 0.108

## Var12 0.627

## Var13 0.859

## Var14 -0.273

## Var15 0.121 0.111 0.343 0.104 0.270 0.475

## Var16 0.253 0.162 0.367 0.106 0.867

## Var17 0.178 0.146 0.812 0.170 0.176

## Var18 0.229 0.151 0.687 0.201 0.106 0.278

##

## Factor1 Factor2 Factor3 Factor4 Factor5 Factor6

## SS loadings 2.260 2.220 1.546 1.473 1.422 1.196

## Proportion Var 0.126 0.123 0.086 0.082 0.079 0.066

## Cumulative Var 0.126 0.249 0.335 0.417 0.496 0.562

##

## Test of the hypothesis that 6 factors are sufficient.

## The chi square statistic is 148.98 on 60 degrees of freedom.

## The p-value is 1.59e-09

Сравнение моделей пятифакторной и шестифакторной моделей

chisq.difference <- EFA5$STATISTIC-EFA6$STATISTIC

df.difference <- EFA5$dof - EFA6$dof

1-pchisq(chisq.difference, df.difference)

## objective

## 0

Построение матицы факторных нагрузок для шестифакторной модели

EFA61 <- factanal(formulaEFA, factors=6, data=Mac.values, rotation = "Varimax")

load3 <- as.data.frame.matrix(EFA61$loadings)

round(load3,2)

## Factor1 Factor2 Factor3 Factor4 Factor5 Factor6

## Var1 0.05 -0.03 -0.05 -0.10 0.03 -0.05

## Var2 0.79 0.10 0.16 0.11 0.10 0.13

## Var3 0.76 0.00 0.06 0.12 0.12 0.11

## Var4 0.80 0.07 0.11 0.13 0.09 0.17

## Var5 0.32 0.07 0.12 0.68 0.07 -0.01

## Var6 0.27 0.11 0.08 0.10 0.13 0.68

## Var7 0.23 0.10 0.09 0.11 0.18 0.55

## Var8 0.37 0.08 0.15 0.24 0.14 0.53

## Var9 0.11 0.08 0.15 0.83 0.10 0.16

## Var10 0.07 0.74 0.15 0.08 0.14 0.09

## Var11 0.13 0.56 0.14 0.09 0.23 0.05

## Var12 0.10 0.62 0.14 0.02 0.04 0.07

## Var13 0.02 0.85 0.12 0.05 0.07 0.04

## Var14 -0.01 -0.07 0.04 -0.27 -0.07 -0.03

## Var15 0.09 0.08 0.52 0.29 0.28 0.07

## Var16 0.14 0.18 0.93 0.11 0.25 0.06

## Var17 0.15 0.13 0.31 0.09 0.79 0.11

## Var18 0.16 0.18 0.39 0.11 0.66 0.14

Построение матрицы нагрузок для пятифакторной модели

EFA51 <- factanal(formulaEFA, factors=5, data=Mac.values, rotation = "Varimax")

load3 <- as.data.frame.matrix(EFA51$loadings)

round(load3,2)

## Factor1 Factor2 Factor3 Factor4 Factor5

## Var1 0.05 0.00 -0.04 -0.11 -0.05

## Var2 0.79 0.17 0.10 0.12 0.12

## Var3 0.76 0.13 0.00 0.11 0.12

## Var4 0.80 0.13 0.07 0.13 0.16

## Var5 0.32 0.12 0.07 0.68 -0.01

## Var6 0.27 0.15 0.11 0.10 0.68

## Var7 0.23 0.19 0.10 0.11 0.56

## Var8 0.37 0.21 0.08 0.24 0.52

## Var9 0.11 0.17 0.08 0.84 0.16

## Var10 0.08 0.20 0.74 0.08 0.09

## Var11 0.13 0.26 0.56 0.09 0.06

## Var12 0.10 0.12 0.62 0.03 0.06

## Var13 0.02 0.12 0.86 0.05 0.04

## Var14 0.00 -0.02 -0.06 -0.26 -0.05

## Var15 0.11 0.55 0.11 0.32 0.03

## Var16 0.18 0.67 0.23 0.19 0.03

## Var17 0.15 0.77 0.13 0.07 0.13

## Var18 0.15 0.80 0.18 0.09 0.14

Альтернативные функции для РФА

psych::fa.parallel(Mac.values)

## Parallel analysis suggests that the number of factors = 5 and the number of components = 5

fa<-psych::fa(Mac.values, 5)

psych::fa.plot(fa, labels=rownames(fa$loadings))



psych::fa.diagram(fa,cut=0.2, simple=F, errors=T)

Конфирматорный анализ. Потсроение пятифакторной модели.

cfa2<- cfa(model=" Quality =~ Var18+Var17+Var16;

AA =~ Var13+Var12+Var11+ Var10;

ICG =~ Var2+Var3+Var4+Var15;

Loality =~ Var9+Var5;

BGС =~ Var6 +Var7+Var8",

data=as.matrix(Mac.values))

cfa2

## lavaan (0.5-23.1097) converged normally after 43 iterations

##

## Number of observations 881

##

## Estimator ML

## Minimum Function Test Statistic 787.669

## Degrees of freedom 94

## P-value (Chi-square) 0.000

library(semPlot)

semPaths(cfa2, whatLabels="est", rotation=2, nCharNodes=0)

Построение статистики согласия

summary(cfa2, fit.measures=TRUE)

## lavaan (0.5-23.1097) converged normally after 43 iterations

##

## Number of observations 881

##

## Estimator ML

## Minimum Function Test Statistic 787.669

## Degrees of freedom 94

## P-value (Chi-square) 0.000

##

## Model test baseline model:

##

## Minimum Function Test Statistic 6164.161

## Degrees of freedom 120

## P-value 0.000

##

## User model versus baseline model:

##

## Comparative Fit Index (CFI) 0.885

## Tucker-Lewis Index (TLI) 0.853

##

## Loglikelihood and Information Criteria:

##

## Loglikelihood user model (H0) -21123.317

## Loglikelihood unrestricted model (H1) -20729.483

##

## Number of free parameters 42

## Akaike (AIC) 42330.634

## Bayesian (BIC) 42531.439

## Sample-size adjusted Bayesian (BIC) 42398.056

##

## Root Mean Square Error of Approximation:

##

## RMSEA 0.092

## 90 Percent Confidence Interval 0.086 0.097

## P-value RMSEA <= 0.05 0.000

##

## Standardized Root Mean Square Residual:

##

## SRMR 0.084

##

## Parameter Estimates:

##

## Information Expected

## Standard Errors Standard

##

## Latent Variables:

## Estimate Std.Err z-value P(>|z|)

## Quality =~

## Var18 1.000

## Var17 0.895 0.036 24.854 0.000

## Var16 0.864 0.040 21.688 0.000

## AA =~

## Var13 1.000

## Var12 0.694 0.038 18.335 0.000

## Var11 1.015 0.054 18.793 0.000

## Var10 0.952 0.041 23.047 0.000

## ICG =~

## Var2 1.000

## Var3 1.004 0.041 24.792 0.000

## Var4 1.123 0.043 26.169 0.000

## Var15 0.460 0.054 8.575 0.000

## Loality =~

## Var9 1.000

## Var5 0.989 0.071 13.905 0.000

## BGС =~

## Var6 1.000

## Var7 0.893 0.058 15.437 0.000

## Var8 1.154 0.067 17.184 0.000

##

## Covariances:

## Estimate Std.Err z-value P(>|z|)

## Quality ~~

## AA 0.511 0.050 10.219 0.000

## ICG 0.453 0.045 9.975 0.000

## Loality 0.442 0.055 8.023 0.000

## BGС 0.457 0.045 10.205 0.000

## AA ~~

## ICG 0.251 0.044 5.707 0.000

## Loality 0.291 0.053 5.445 0.000

## BGС 0.297 0.043 6.954 0.000

## ICG ~~

## Loality 0.549 0.057 9.677 0.000

## BGС 0.558 0.047 11.930 0.000

## Loality ~~

## BGС 0.435 0.052 8.408 0.000

##

## Variances:

## Estimate Std.Err z-value P(>|z|)

## .Var18 0.333 0.035 9.509 0.000

## .Var17 0.485 0.035 14.037 0.000

## .Var16 0.853 0.048 17.647 0.000

## .Var13 0.551 0.044 12.392 0.000

## .Var12 0.834 0.045 18.376 0.000

## .Var11 1.654 0.091 18.161 0.000

## .Var10 0.577 0.043 13.445 0.000

## .Var2 0.445 0.033 13.564 0.000

## .Var3 0.628 0.040 15.751 0.000

## .Var4 0.573 0.042 13.719 0.000

## .Var15 2.077 0.100 20.674 0.000

## .Var9 0.940 0.094 10.017 0.000

## .Var5 0.565 0.085 6.639 0.000

## .Var6 0.742 0.050 14.971 0.000

## .Var7 0.910 0.053 17.156 0.000

## .Var8 0.791 0.059 13.497 0.000

## Quality 1.101 0.073 14.988 0.000

## AA 1.156 0.085 13.608 0.000

## ICG 0.995 0.070 14.241 0.000

## Loality 1.273 0.126 10.090 0.000

## BGС 0.751 0.071 10.606 0.000

Построение статистики модификации коэффициентов.

modindices(cfa2, minimum.value = 10, sort=TRUE)

## lhs op rhs mi epc sepc.lv sepc.all sepc.nox

## 55 Quality =~ Var15 223.105 0.860 0.902 0.597 0.597

## 148 Var16 ~~ Var15 128.304 0.547 0.547 0.280 0.280

## 95 Loality =~ Var15 92.394 0.569 0.642 0.425 0.425

## 80 ICG =~ Var9 51.002 -0.841 -0.838 -0.564 -0.564

## 81 ICG =~ Var5 51.000 0.832 0.830 0.616 0.616

## 67 AA =~ Var15 42.870 0.337 0.362 0.239 0.239

## 109 BGС =~ Var15 41.058 0.613 0.532 0.352 0.352

## 111 BGС =~ Var5 40.565 -0.724 -0.628 -0.466 -0.466

## 110 BGС =~ Var9 40.564 0.732 0.634 0.426 0.426

## 154 Var13 ~~ Var12 40.444 0.245 0.245 0.159 0.159

## 217 Var15 ~~ Var9 36.849 0.338 0.338 0.150 0.150

## 216 Var4 ~~ Var8 36.141 0.193 0.193 0.107 0.107

## 50 Quality =~ Var11 27.747 0.296 0.311 0.184 0.184

## 167 Var12 ~~ Var10 25.080 -0.185 -0.185 -0.123 -0.123

## 229 Var6 ~~ Var7 25.044 0.219 0.219 0.146 0.146

## 112 Var18 ~~ Var17 23.159 0.271 0.271 0.194 0.194

## 210 Var3 ~~ Var8 21.882 -0.148 -0.148 -0.087 -0.087

## 48 Quality =~ Var13 21.817 -0.188 -0.198 -0.151 -0.151

## 76 ICG =~ Var13 20.935 -0.167 -0.166 -0.127 -0.127

## 225 Var9 ~~ Var8 20.511 0.185 0.185 0.093 0.093

## 102 BGС =~ Var13 20.380 -0.209 -0.181 -0.139 -0.139

## 84 ICG =~ Var8 18.666 0.327 0.327 0.244 0.244

## 63 AA =~ Var16 18.641 0.176 0.189 0.146 0.146

## 209 Var3 ~~ Var7 18.413 0.133 0.133 0.085 0.085

## 98 Loality =~ Var8 17.988 0.211 0.238 0.178 0.178

## 155 Var13 ~~ Var11 16.573 -0.227 -0.227 -0.103 -0.103

## 87 Loality =~ Var16 15.901 0.151 0.171 0.132 0.132

## 231 Var7 ~~ Var8 14.931 -0.194 -0.194 -0.118 -0.118

## 78 ICG =~ Var11 13.399 0.191 0.191 0.113 0.113

## 65 AA =~ Var3 12.741 -0.117 -0.126 -0.099 -0.099

## 57 Quality =~ Var5 11.749 -0.213 -0.224 -0.166 -0.166

## 56 Quality =~ Var9 11.749 0.215 0.226 0.152 0.152

## 212 Var4 ~~ Var9 11.587 -0.119 -0.119 -0.059 -0.059

## 104 BGС =~ Var11 11.036 0.219 0.190 0.112 0.112

Построение модели с учетом ковариляции остатков

cfa2.1<- cfa(model=" Quality =~ Var18+Var17+Var16;

AA =~ Var13+Var12+Var11+ Var10;

ICG =~ Var2+Var3+Var4+Var15;

Loality =~ Var9+Var5+Var14;

BGС =~ Var6 +Var7+Var8;

Var13 ~~ Var11;

Var13 ~~ Var12;

Var7 ~~ Var8;

Var4 ~~ Var9;

Var16 ~~ Var15;

Var3 ~~ Var7;

Var9 ~~ Var8;

Var12 ~~ Var10",

data=as.matrix(Mac.values))

cfa2.1

## lavaan (0.5-23.1097) converged normally after 51 iterations

##

## Number of observations 881

##

## Estimator ML

## Minimum Function Test Statistic 536.617

## Degrees of freedom 101

## P-value (Chi-square) 0.000

summary(cfa2.1, fit.measures=TRUE)

## lavaan (0.5-23.1097) converged normally after 51 iterations

##

## Number of observations 881

##

## Estimator ML

## Minimum Function Test Statistic 536.617

## Degrees of freedom 101

## P-value (Chi-square) 0.000

##

## Model test baseline model:

##

## Minimum Function Test Statistic 6238.073

## Degrees of freedom 136

## P-value 0.000

##

## User model versus baseline model:

##

## Comparative Fit Index (CFI) 0.929

## Tucker-Lewis Index (TLI) 0.904

##

## Loglikelihood and Information Criteria:

##

## Loglikelihood user model (H0) -22601.929

## Loglikelihood unrestricted model (H1) -22333.621

##

## Number of free parameters 52

## Akaike (AIC) 45307.859

## Bayesian (BIC) 45556.474

## Sample-size adjusted Bayesian (BIC) 45391.333

##

## Root Mean Square Error of Approximation:

##

## RMSEA 0.070

## 90 Percent Confidence Interval 0.064 0.076

## P-value RMSEA <= 0.05 0.000

##

## Standardized Root Mean Square Residual:

##

## SRMR 0.079

##

## Parameter Estimates:

##

## Information Expected

## Standard Errors Standard

##

## Latent Variables:

## Estimate Std.Err z-value P(>|z|)

## Quality =~

## Var18 1.000

## Var17 0.888 0.038 23.636 0.000

## Var16 0.661 0.036 18.369 0.000

## AA =~

## Var13 1.000

## Var12 0.675 0.054 12.485 0.000

## Var11 1.103 0.061 18.077 0.000

## Var10 1.041 0.069 14.985 0.000

## ICG =~

## Var2 1.000

## Var3 1.018 0.041 24.724 0.000

## Var4 1.133 0.044 26.045 0.000

## Var15 0.354 0.049 7.170 0.000

## Loality =~

## Var9 1.000

## Var5 0.995 0.069 14.455 0.000

## Var14 -0.353 0.053 -6.715 0.000

## BGС =~

## Var6 1.000

## Var7 1.019 0.075 13.503 0.000

## Var8 1.269 0.085 14.887 0.000

##

## Covariances:

## Estimate Std.Err z-value P(>|z|)

## .Var13 ~~

## .Var11 -0.053 0.072 -0.739 0.460

## .Var12 0.182 0.048 3.771 0.000

## .Var7 ~~

## .Var8 -0.185 0.052 -3.530 0.000

## .Var4 ~~

## .Var9 -0.130 0.034 -3.785 0.000

## .Var16 ~~

## .Var15 0.672 0.054 12.377 0.000

## .Var3 ~~

## .Var7 0.109 0.032 3.427 0.001

## .Var9 ~~

## .Var8 0.198 0.042 4.779 0.000

## .Var12 ~~

## .Var10 -0.052 0.052 -0.997 0.319

## Quality ~~

## AA 0.498 0.051 9.857 0.000

## ICG 0.457 0.046 10.031 0.000

## Loality 0.420 0.054 7.735 0.000

## BGС 0.429 0.044 9.649 0.000

## AA ~~

## ICG 0.244 0.042 5.753 0.000

## Loality 0.293 0.052 5.679 0.000

## BGС 0.266 0.040 6.714 0.000

## ICG ~~

## Loality 0.546 0.056 9.707 0.000

## BGС 0.513 0.047 10.963 0.000

## Loality ~~

## BGС 0.386 0.049 7.837 0.000

##

## Variances:

## Estimate Std.Err z-value P(>|z|)

## .Var18 0.324 0.038 8.522 0.000

## .Var17 0.492 0.036 13.492 0.000

## .Var16 0.944 0.049 19.179 0.000

## .Var13 0.659 0.071 9.231 0.000

## .Var12 0.913 0.065 13.959 0.000

## .Var11 1.571 0.112 14.016 0.000

## .Var10 0.490 0.070 7.028 0.000

## .Var2 0.459 0.033 13.875 0.000

## .Var3 0.632 0.040 15.727 0.000

## .Var4 0.552 0.041 13.311 0.000

## .Var15 2.095 0.101 20.811 0.000

## .Var9 0.949 0.090 10.509 0.000

## .Var5 0.561 0.081 6.919 0.000

## .Var14 2.272 0.110 20.626 0.000

## .Var6 0.800 0.053 15.208 0.000

## .Var7 0.802 0.064 12.538 0.000

## .Var8 0.658 0.073 8.969 0.000

## Quality 1.110 0.075 14.788 0.000

## AA 1.047 0.099 10.621 0.000

## ICG 0.980 0.070 14.086 0.000

## Loality 1.264 0.123 10.284 0.000

## BGС 0.693 0.070 9.878 0.000

Сравнение моеделй.

lavTestLRT(cfa2, cfa2.1)

## Warning in lavTestLRT(cfa2, cfa2.1): lavaan WARNING: some models are based

## on a different set of observed variables

## Chi Square Difference Test

##

## Df AIC BIC Chisq Chisq diff Df diff Pr(>Chisq)

## cfa2 94 42331 42531 787.67

## cfa2.1 101 45308 45556 536.62 -251.05 7 1

Построение пятифакторной модели для конфирматорного факторного анализа

semPaths(cfa2.1, nCharNodes=0)

...