A Match Made in Heaven? The Effect of Congruency Between Accent and Promoted
Product in Radio Adverts

A Match Made in Heaven? The Effect of Congruency Between Accent and Promoted
Product in Radio Adverts

Samantha Trow

2018

Research consistently shows that accents are powerful social cues used in our
everyday interactions as well as in advertisements; they can change how we perceive
others and potentially also associated products or brands. Recent studies have
explored the effect of congruency between the accent of the speaker in adverts and the
country-of-origin of the advertised products. Yet the findings from research on the
congruency effect is mixed and sparse. Therefore, this study investigated further into
the effect of congruency. Participants were randomly assigned to one of four
experimental conditions. The study employed a 2 (Accent: Northern English vs.
Italian) x 2 (Product: fish and chips vs. pizza) between participant design. In doing
this, two adverts had a congruent accent-product pair (e.g., Northern English speaker
advertising a fish and chips brand) and two ads were accent-product incongruent (e.g.,
Northern English speaker advertising a pizza brand). After listening to the ads,
participants then completed a questionnaire which measured participants’ brand
memory, attention to the ad, purchase intentions, perceived similarity to the speaker
and evaluations of the brand, advert and speaker. The results showed no congruency
effect, however other striking findings were revealed that will be discussed in this
paper.

None

This study used a 2 (Accent: Northern English vs. Italian) x 2 (Product: fish and chips
vs. pizza) between subject design. The dependant variables were participants’
attention to the ad, memorability of the advertised brand name, purchase intentions,
evaluations of the speaker, and attitude towards the ad and brand. Additionally, the
evaluations of the speaker included their perceived warmth, competence, sociointellectual status, aesthetic qualities, and dynamism traits.
Participants
Through opportunity sampling, 82 participants were recruited. This sample
comprised of 29 males and 53 females. Participants were randomly assigned to one of
the four conditions. The participants’ age ranged from 19 to 65 (Mage = 25.5 years,
SDage = 10.8). All but one participant were native speakers of English.
Materials
Radio Advertisements. For this experiment, four radio adverts were created
(see Appendix A). Two ads were accent-product congruent (Italian accent and pizza,
Northern English accent and fish and chips) and two ads were accent-product
incongruent (Italian accent and fish and chips, Northern English accent and pizza). In
order to create the adverts, two male speakers were recruited. One of the speakers
spoke with an authentic Northern English accent and one of the speakers spoke with
an authentic Italian accent, both spoke at similar paces with no major differences in
their tone of voice.
Questionnaire. The questionnaire used in the experiment was created via the
survey software, Qualtrics. The questionnaire took approximately 10 minutes to
complete. The items and scales used to measure the dependent variables are discussed
below.
Brand attitude. Participants’ attitude towards the advertised brand was
measured using a 4-item, 7-point bipolar scale used in Liu, Wen, Wei. and Zhao’s
(2013) study (ɑ = .92). See Appendix B for the full subscale.
Ad attitude. Participants’ attitude towards the advert subscale was taken from
Lalwani, Lwin, and Li’s (2005) study. The participants were asked to rate the radio
advert using 4-items with 7-point bipolar scales (ɑ = .87). See Appendix C.
Attention to the ad. Also taken from Lalwani et al.’s (2005) study, were 3-
items with 7-point likert scales to measure participants’ attention to the ad (ɑ = .24).
The Cronbach’s alpha score was low however removing items did not increase the
alpha significantly to represent a robust measure. See Appendix D.
Purchase intentions. In addition, based on the scales used in Hornikx, van
Meurs, and Hof’s (2013) research, the questionnaire included 3-items with 7-point
bipolar scales to measure participants’ purchase intentions (ɑ = .88). See Appendix E.
Competence and warmth. The questionnaire included questions which
measured the perceived competence and warmth of the speaker. The 9-items for
competence (ɑ = .90) and 9-items for warmth (ɑ = .92) were presented together. The
scale used for the items were 7-point likert scales (1= Strongly Disagree, 7= Strongly
Agree), taken from Rudman and Glick’s (1999) study. The list of items used can be
found in Appendix F and G, respectively.
Socio-intellectual status, aestheticism and dynamism. Also, the questionnaire
included the Speech Dialect Attitudinal Scale by Mulac (1975, 1976). This consisted
of 12-items (four items for each subscale) with 7-point bipolar scales measuring the
participants’ perceived socio-intellectual status (ɑ = .85), aestheticism (ɑ = .85), and
dynamism of the speaker (ɑ = .76). See Appendix H.
Similarity. To measure participants’ perceived similarity to the speaker in the
ad, the questionnaire included 3-items with 7-point likert scales (ɑ = .80) taken from
Lalwani et al.’s (2005) questionnaire. See Appendix I.
Manipulation check. The questionnaire examined if participants correctly
identified the accent used by the speaker in the ad. Participants were asked “What was
the accent of the speaker in the ad?”.
Memorability of the brand name. At the end of the questionnaire the
participants were asked the open-ended question “Please write down the product’s
brand name that was advertised in the radio ad you listened to.”.
Additional questions. The questionnaire included additional questions which
investigated whether any factor other than accent affected participants’ responses.
These questions consisted of 7-point bipolar scales, 7-point likert scales, unipolar
scales, and open-ended questions (see Appendix J). The questions measured the
comprehensibility of the speaker in the ad, participants’ attitudes towards the accent,
accent of the participant, likability of the advertised products, hunger, and native
language of the participant. The questionnaire also asked demographic questions.
Procedure
After giving the informed consent, participants were randomly assigned to an
experimental condition and sent the link to the Qualtrics questionnaire. At the
beginning of the questionnaire the radio ad was played followed by the questions. The
order in which the items were presented were brand attitude, ad attitude, attention to
ad, purchase intentions, warmth, and competence, socio-intellectual status of speaker,
aestheticism of speaker, dynamism of speaker, similarity to speaker,
comprehensibility of speaker, accent of the speaker, attitude towards the ad, accent of
the participant, likeability of the advertised product, frequency of eating advertised
product, hunger of participant, participants’ first language, brand name memorability,
and finally followed by the demographic questions. On completion of the
questionnaire, participants were thanked and debriefed.
Analysis
A multivariate ANOVA was used to test the main and interaction effects of
accent and product on participants’ evaluations. Also, separate univariate ANOVAs
were conducted to explore if there were any covariate effects on participants attention
to the ad, brand memorability, evaluations of brand, ad or speaker. The covariate
variables were participants’ perceived similarity to the speaker, comprehensibility of
the speaker, participants’ attitude towards the speaker’s accent, hunger, frequency of,
and likability of eating the advertised product.

Lancaster University

Data/SPSS.sav

Trow2018

Rebecca James

Open

None

English

Data

LA1 4YF

Dr Tamara Rakić

MSc

Advertising, Marketing, Cognitive Perception

82 participants

MANOVA, ANOVA