香京julia种子在线播放

Advertising literacy, also referred to as persuasion knowledge (Friestad and Wright, 1994), includes a wide range of knowledge, attitudes, and skills needed to critically process advertising. Dispositional advertising literacy (i.e., knowledge and critical attitudes that are in the child’s mind regardless of exposure to advertising; Hudders et al., 2017) can be defined in at least two dimensions: conceptual and attitudinal advertising literacy. Dispositional conceptual advertising literacy includes, for instance, the understanding of advertising’s selling intent (John, 1999) and persuasive intent (Moses and Baldwin, 2005). Dispositional attitudinal advertising literacy includes a general disliking and skeptical attitude toward advertising (Rozendaal et al., 2011b). Dispositional advertising literacy gradually increases with age (e.g., John, 1999; Moses and Baldwin, 2005; Rozendaal et al., 2011b). It develops from simple to more sophisticated knowledge and beliefs about the nature and working of advertising (Wright et al., 2005; Rozendaal et al., 2011b; Hudders et al., 2017). The increase in children’s dispositional advertising literacy depends on both the development of children’s cognitive and social abilities, including information processing and perspective taking skills, and the socialization of children as consumers (John, 1999). That is, through experience in coping with advertising, children acquire advertising-related knowledge and attitudes that are relevant to their functioning as consumers (Moses and Baldwin, 2005).

To investigate if a disclosure can stimulate children to activate their dispositional advertising literacy when they are exposed to in-vlog advertising, the present study draws on insights from information-processing theory (Atkinson and Shiffrin, 1968) and schema theory (Anderson, 1978). Information-processing theory states that, rather than directly responding to incoming information, people first process this information and then respond accordingly (Atkinson and Shiffrin, 1968). Incoming information is first represented and stored in short-term memory, which only has limited capacity. In order for the information to be retained and stored for later use, it needs to be transferred to long-term-memory (Atkinson and Shiffrin, 1968; Roedder, 1981). New, incoming information can then serve as a cue to retrieve this stored information. Subsequently, the incoming information is processed in relation to the retrieved information that was already stored in long-term memory (Atkinson and Shiffrin, 1968; Roedder, 1981; Rozendaal et al., 2011b).

The information that is stored in long-term memory is believed to be organized in mental structures called schemas (Aronson et al., 2005). Schemas are knowledge structures that hold all information we have on a certain topic, for instance about advertising. Each schema consists of all relevant information associated with that topic and usually contains several subschemas (Anderson, 1978). In the case of the advertising schema, it is believed that there is one overarching schema that includes all general knowledge and beliefs about advertising (Dahlén and Edenius, 2007), with separate subschemas for dispositional conceptual and attitudinal advertising literacy (Friestad and Wright, 1994; Dahlén and Edenius, 2007; Evans and Park, 2015).

When combining the perspectives of information-processing and schema theories, they predict that exposure to advertising, which serves as a cue for information retrieval, should result in the activation of the advertising schema, including the subschema containing conceptual and attitudinal advertising literacy. As such, exposure to advertising could lead to the activation of children’s dispositional advertising literacy. However, information-processing theory also predicts that activating relevant information can be difficult for children. Even though information retrieval as a results of exposure to media content is largely automatic (Robinson and Neighbors, 2006), children also need to be able to use retrieval strategies (Roedder, 1981) in order to correctly use and link the retrieved information.

The activation of advertising literacy can be stimulated by exposing children to a ‘retrieval’ cue (Roedder, 1981; Hudders et al., 2017). Such a cue (i.e., incoming information) can result in the retrieval of related information (e.g., dispositional advertising literacy as stored in the advertising schema). A cue to stimulate children to activate their advertising literacy can, for instance, be a sponsorship disclosure, which is a forewarning about the persuasive intent of the message (Boerman et al., 2012; Rozendaal et al., 2016a). Earlier research has shown that disclosures can help children better recognize product placement as a type of advertising (De Jans et al., 2018). Furthermore, studies have shown that a visual disclosure can be helpful in activating children’s advertising literacy when it comes to embedded advertising in television programs and movies (e.g., De Pauw et al., 2017; Van Reijmersdal et al., 2017). For instance, a study by De Pauw et al. (2017) showed that children who were exposed to a visual warning cue (i.e., disclosure) before watching a movie clip with brand placement displayed higher levels of conceptual advertising literacy than children who were exposed to an auditory warning cue. Similar, a study by De Jans et al. (2019a) showed that including a disclosure in a vlog that contained advertising increased children’s attitudinal advertising literacy. In line with the results from previous studies, we set up our first hypothesis:

In order for children to activate their conceptual and attitudinal advertising literacy, these types of advertising literacy have to be present in the first place. When children do not have certain knowledge about the persuasive and selling intent of advertising in general (i.e., dispositional conceptual advertising literacy) or do not have a general skeptical attitude toward advertising (i.e., dispositional attitudinal advertising literacy), they will also not be able to activate this knowledge or attitude when they are exposed to advertising. Therefore, we argue that there will be an interaction effect between exposure to the disclosure and children’s dispositional advertising literacy. Specifically, we expect that the effect of the disclosure on advertising literacy activation will be stronger for children with higher levels of dispositional advertising literacy. This leads to the second hypothesis:

Insights in children’s cognitive development suggest that activating relevant information from long-term memory is more difficult for younger children (i.e., younger than 12 years) than for older children (i.e., 12 years and older). As such, the processing of in-vlog advertising can differ markedly between younger and older children (Hudders et al., 2017).

The prefrontal brain plays a significant role in children’s ability to activate and retrieve relevant information from memory (Uytun, 2018) and doesn’t mature fully until late adolescence (Casey et al., 2005). However, from the age of 12 onwards, most children are able to use retrieval strategies to activate relevant information from memory and use it efficiently (i.e., strategic processors; Roedder, 1981; John, 1999), for instance, to use activated advertising literacy to counter advertising effects. Children younger than 12 years old are also able to do this, but only when they are prompted or cued (i.e., cued processors; Roedder, 1981; Rozendaal et al., 2011b). This implicates that, even though younger children may have a fairly high level of dispositional conceptual and attitudinal advertising literacy, activating this literacy and using it to decrease susceptibility may still be difficult for them (John, 1999). Children in this younger age group must be cued to activate their advertising literacy. Therefore, use of a disclosure is thus expected to be more important for activating advertising literacy in the younger age group than it is in the older age group. Thus, age was included as a moderating variable in this study. More specifically, we expect that age moderates the effect of the disclosure as hypothesized in H2. This leads to the following hypothesis:

To investigate whether and how advertising literacy activation is related to children’s responses to the advertised brand (i.e., brand attitude), the present study draws on insights from the Persuasion Knowledge Model (PKM, Friestad and Wright, 1994) and reactance theory (Brehm and Brehm, 1981). The Persuasion Knowledge Model (Friestad and Wright, 1994) suggests that higher levels of advertising literacy activation can change the way children respond to an advertising message and its source (i.e., the brand). In the context of this study this means that when children become aware of the fact that a vlog is sponsored, they may realize that the vlog is not just entertaining or informative but is meant to persuade. This can trigger feelings about the honesty, trustworthiness, and credibility of the message. According to Brehm and Brehm (1981) reactance theory, people do not want to be manipulated and desire to maintain the freedom to feel and think what they want. Therefore, when advertising literacy schemas are active, people are assumed to use the activated knowledge and attitudes to cope with the persuasion attempt, which can result in negative responses (e.g., more negative brand attitudes).

Earlier research has found that, among adults, a better understanding of sponsored content resulted in a more critical attitude toward not only the sponsored content but also toward the brands (Boerman et al., 2014). However, the empirical evidence among children for this relationship is less conclusive. For example, with regard to conceptual advertising literacy, only few studies (De Jans et al., 2017) show that stronger activation of this concept leads to more negative brand responses. Interestingly, most studies report positive effects of conceptual advertising literacy activation on children’s brand responses (e.g., Rozendaal et al., 2009; Van Reijmersdal et al., 2015; Vanwesenbeeck et al., 2016, 2017; De Pauw et al., 2017). One explanation for the positive effect could be that children’s realization that the message is made to persuade (i.e., conceptual advertising literacy) actually persuades them to like, want, or buy the advertised product. In this study, we specifically focus on brand attitude as a brand response because it is an important indicator of how a brand is perceived and can predict people’s behavior toward the brand (Brown and Stayman, 1992). Based on the existing empirical evidence we hypothesize the following:

Earlier research on the relationship between attitudinal advertising literacy activation and children’s brand responses provides more convincing evidence for the defense view. These studies all show that children’s attitudinal advertising literacy has a negative effect on brand responses (e.g., Rozendaal et al., 2013, 2016a; Vanwesenbeeck et al., 2016, 2017). Thus, in contrast to conceptual advertising literacy, attitudinal advertising does seem to negatively influence brand responses, as suggested by the PKM (Friestad and Wright, 1994). An explanation for this could be that the negative attitude children have toward the specific advertising message, or toward advertising in general, spills over to the brand, leading to more negative brand responses. This is also referred to as affect transfer (Machleit and Wilson, 1988) and led to the following hypothesis:

The conceptual models for the hypotheses can be found in Figure 1 (conceptual advertising literacy) and Figure 2 (attitudinal advertising literacy).

Over the last 10 years, several scholars have worked on the development of a valid and reliable measurement tool to assess children’s dispositional advertising literacy (e.g., D’Alessio et al., 2009; Ham et al., 2015; Rozendaal et al., 2016b). These tools aim to assess children’s advertising literacy that is present in the child’s intellect regardless of advertising exposure (i.e., knowledge and attitudes toward advertising in general; Hudders et al., 2017). In empirical studies in which children are exposed to specific advertisements to investigate their responses to those advertisements, these dispositional measures are often used in an adapted form to measure children’s advertising literacy in relation to those specific advertisements (i.e., situational advertising literacy; Hudders et al., 2017). For example, items such as “Is the purpose of advertising to sell products?” are adapted into “Is the purpose of this advertisement to sell products?” This way, researchers aim to measure whether children activate their general knowledge of advertising (i.e., dispositional advertising literacy) when they are confronted with a specific advertising message.

An issue with measuring children’s advertising literacy activation in this manner is that it is based on self-report questionnaires. Although questionnaires are suitable for assessing the dispositional dimensions of advertising literacy, they are considered less appropriate for assessing advertising literacy activation (Rozendaal et al., 2016b). There are several reasons for this. First, children have to retrieve information from memory in order to answer the questions in the questionnaire. This can be difficult for children, because this means that they have to introspectively search for the answers (Dunham et al., 2006). With respect to advertising literacy activation, retrieving information from memory is especially hard, because the questions would relate to the processing of advertising, which is usually automatic and subconscious (Vandeberg et al., 2013). Second, the wording of the questions and answering options may prompt the answers that children give (Brace, 2004). In this case it could mean that children choose the option that is most appealing to them, or that is in line with what they think the researcher wants to know. Third, and most importantly, questionnaires stimulate respondents to consciously and elaborately think about the processing of advertising. As a consequence, questionnaires may activate post hoc rationalizations (Vandeberg, 2014) that do not reflect the cognitive and affective processes that were actually activated during exposure to the advertising message. Concluding, the self-report measurement tools that are currently used to assess children’s dispositional advertising literacy have several disadvantages that make them less suitable to assess children’s advertising literacy activation.

An alternative way in which children’s activation of advertising literacy can be measured, and a way with which the disadvantages of self-report can be avoided, is the use of indirect measures (e.g., reaction time measures). The use of indirect measures is very common in schema activation research (e.g., Kim and Hancock, 2016; Leavitt et al., 2016). One of the most important advantages of indirect measures is that they do not require people to consciously reflect on a past experience (Vandeberg et al., 2013). Therefore, they are better able to capture subconscious processes (De Houwer, 2006), such as the activation of dispositional advertising literacy as a response to advertising exposure. Although the use of indirect measures with children is not yet very widespread, previous studies have shown that it is possible to use these types of reaction time measures when doing research with children aged 6 years and older (e.g., Baron and Banaji, 2006; Banse et al., 2010; Cvencek et al., 2011).

In the current study, in addition to a self-reported measure, a new indirect measurement task was used to assess children’s advertising literacy activation: The Advertising Literacy Activation Task (ALAT; Hoek et al., 2019). The ALAT is based on the idea that when a certain concept (i.e., schema) is activated, it is easier to process and categorize words related to this concept (i.e., schema). A concept is usually active in relation to a real-life situation or condition (e.g., being in a supermarket activates thoughts about supermarkets; Robinson and Neighbors, 2006), but can also be activated by exposure to media content (Robinson and Neighbors, 2006). For example, exposure to an advertising message can activate thoughts about advertising (i.e., the advertising schema, including dispositional conceptual and attitudinal advertising literacy). The extent to which a schema is activated can be derived from the speed and accuracy with which a person can place related and unrelated words into a relevant category (Fazio and Olson, 2003; Robinson and Neighbors, 2006). This has been shown in several categorization tasks within priming research (e.g., Fazio et al., 1986; Zeelenberg et al., 2003; Ortells et al., 2016). For instance, Fazio et al. (1986) showed that people are faster in categorizing positive words when they were presented with a positive prime prior to categorizing the positive word.

In the ALAT, the categorization task is designed in such a way that children are asked to categorize targets, in this case advertising-related words (e.g., buy), non-advertising related words (e.g., weather) and attitudinal words (e.g., boring) into one of two categories: advertising or news. The ALAT is an indirect measurement task, because it does not directly ask participants questions regarding their advertising literacy activation, but indirectly assesses it from their reactions on the task (Fazio and Olson, 2003). The two most important premises of the ALAT are that (1) exposure to an advertising cue (e.g., an advertisement and/or sponsorship disclosure) activates dispositional conceptual and attitudinal advertising literacy, which (2) in turn increases accuracy and speed of categorizing words associated with advertising.

To explore if the way in which advertising literacy activation is measured (i.e., indirect measurement vs. self-report) affects the hypothesized relationships as depicted in Figures 1, 2, we formulated the following research question:

In total, 289 participants took part in the experiment. Thirty-one participants were excluded from the data analysis for one of the following reasons: (1) They did not sufficiently master the Dutch language, (2) The manipulation failed (e.g., the participant saw both versions of the stimulus material), (3) They did not finish the entire experiment, (4) They did not fall within the age range for this study (i.e., younger than 7 years old or older than 16 years old), or (5) Their performance on the indirect measurement task was not good enough (i.e., more than 25% of their data was removed after outlier analysis). Of the participants that remained (N = 258), 50.0% were girls. The participants’ mean age was 10.45 (SD = 1.94, range: 7–16). Almost two-thirds of the participants (65.5%) were in primary school, and 34.5% of the participants were in secondary school.

Data collection took place in the science museum, which has a designated area where scholars are allowed to do research. This research area functioned as a research lab where the experiment was conducted under controlled conditions. The research area consists of two rooms: one room where participants and their parents were given details about the study and parents signed the consent form and one room where the children actually participated in the study. In the second room, four desks were each equipped with a laptop, a headphone, a mouse, and a button box. Recruitment of the participants took place inside the museum. Participants were told that the study was about vlogs and the way that people process information presented in vlogs. When they (and their parents) agreed to participate, they were taken to the area where the study took place. The consent form was signed by a parent or guardian for all children. After signing the consent form, participants were taken to the room where the actual study was conducted. They were positioned behind one of the laptops and asked to put on the headphone.

In the first part of the study, participants watched a vlog that contained either a disclosure at the beginning of the vlog (experimental condition) or not (control condition). Next, they performed the indirect measurement task to assess their advertising literacy activation. Finally they filled out a questionnaire. The experiment took approximately 20 min to complete. After completion, participants were thanked for their contribution and given a debriefing letter which explained the purpose of the study. None of the participants received an incentive for participating in the study.

The vlog used in this study was selected on the basis of three selection criteria. First, the vlog needed to be attractive for children of all ages (e.g., we did not want to use a vlog that would be perceived as being childish). Second, the product/brand in the vlog needed to be attractive for children of all ages. Third, the product was preferably easy to buy for the participants (e.g., not too expensive and easily available). Based on these criteria, we selected a video of a Dutch male vlogger, who has approximately 146,000 followers on YouTube. In the vlog, he visits a popular pizza chain to make pizzas for a (sponsored) pizza lunch at a secondary school. He then visits the school where the children have to complete a set of challenges in order to win the lunch. The original vlog lasted 7 min and 3 s. For the purpose of this study some parts of the vlog were removed (e.g., a fragment in which the vlogger drives to the school, which was not necessary for the narrative). The final version of the vlog used in this study lasted 4 min and 36 s.

The selected vlog includes several features which are typical for (sponsored) vlogs. First, the vlogger films himself using two main camera angles: one where the camera is focused on the environment (e.g., the vloggers’ house or a shopping street) and another that’s selfie-style. Second, the vlogger films himself in ordinary settings (e.g., on the street, in a pizza store, in a class-room) while constantly providing engaging, improvised commentary as he films. Third, the vlog shows a give-away, which is an persuasive tactic that is often used in sponsored vlogs. The give-away consists of winning a pizza lunch from the sponsoring pizza chain for the entire class.

From one of our earlier studies (Van Berlo et al., 2017), in which we extensively pretested stimulus materials including brand placement for a similar age group, we know that ‘pizza’ is a highly relevant product category for children and teenagers. With regard to the selected pizza chain brand, this study showed that most children were familiar with the brand and held neutral attitudes toward it (important to prevent a ceiling effect on brand attitude).

For the experimental condition, we added a disclosure at the start of the video because a disclosure at the beginning of a video is more effective than a disclosure concurrent with the product placement (De Pauw et al., 2017). The disclosure was portrayed in white letters on a black screen, reading “[name vlogger] has been paid by [name company] to advertise in this video” (Van Reijmersdal et al., 2020). The disclosure was based on social media advertising codes (FTC, 2013; WOMMA, 2013) and was shown for 10 s. In the control condition, we added a black screen without any message at the beginning of the video. Like the disclosure message (which was portrayed in white letters on a black screen), the black screen was shown for 10 s. In doing so, we kept the length of the two videos and the time between the start of the video (i.e., the moment when the participant clicks the start button) and the actual start of the vlog content (i.e., when the vlogger comes into the picture) exactly the same. Thus, participants in both conditions saw a black screen during the first 10 seconds of the video. The only difference was that in the experimental condition, the disclosure text was added to the black screen.

We included two questions to check if participants perceived the manipulation as intended. First, we asked them if they saw a text at the beginning of the video (answer options: yes; no; I don’t know). A Chi² analysis showed that the manipulation was successful. In the condition with the disclosure, 78.0% of the children reported seeing a disclosure versus 14.3% of the control condition, χ²(2, N = 258) = 107.78, p < 0.001. We then asked them which text they saw. There were five answer options. One answer option was correct (“[name vlogger] is being paid by [name company] to advertise in this video”). Three answer options were incorrect (1) “This is a video from YouTube,” (2) “[name vlogger] hopes that you like this video, click on the thumps up,” and (3) “YouTube earns money by airing this video”). There was also one option mentioning not seeing the text. Of the children in the disclosure condition who indicated that they saw a disclosure, 70.9% identified the correct disclosure text. Of the children in the control condition, 6.2% indicated to have seen the disclosure text that was shown in de experimental condition (which they did not see). Most participants in the control condition indicated correctly that they had not seen any text (68,2%).

All analyses were done in SPSS (Version 25). We first performed several randomization checks to see which variables needed to be included as covariates in the main analyses. Randomization checks were all χ² and t-test analyses. Second, we tested the hypotheses by doing two moderated mediation analyses. Both analyses were done with brand attitude as the outcome variable and condition (i.e., exposure to the disclosure) as the predictor variable. The first analysis tested the model presented in Figure 1 with dispositional conceptual advertising literacy as the first moderator variable. Age was included as a second moderator. For this, age was recoded into two categories: 7- to 11-year-olds (i.e., cued processors Roedder, 1981; n = 181; M = 9.45, SD = 1.21; 52% were girls) and 12- to 16-year-olds (i.e., strategic processors; Roedder, 1981; n = 77; M = 12.82, SD = 1.10; 46% were girls). This analysis had three mediator variables. Two of the mediators were assessed with the indirect measurement task and meant to indicate conceptual advertising literacy activation (difference scores for accuracy of categorizing the conceptual words and speed of categorizing the conceptual words). We further included conceptual advertising literacy activation as assessed by the self-reported questionnaire as a mediator. The three mediators were tested in parallel.

The second analysis tested the model presented in Figure 2 with dispositional attitudinal advertising literacy as the first moderator variable and age group as the second moderator. This analysis also had three mediator variables. Two of the mediators were assessed with the indirect measurement task and meant to indicate attitudinal advertising literacy activation (difference scores for accuracy of categorizing the attitudinal words and speed of categorizing the attitudinal words). We further included attitudinal advertising literacy activation as assessed by the self-reported questionnaire as a mediator. The three mediators were tested in parallel. Both models require moderated mediation analysis—we used Hayes (2017) SPSS PROCESS macro version 3.2.01, model 11 to test the hypotheses simultaneously. The default bootstrapping procedure was applied with 95% bias corrected accelerated confidence intervals based on 10,000 bootstrap samples.

We ran a number of randomization checks to see which variables needed to be included as covariates in the main analyses. The following variables were equally distributed across both conditions: sex, χ²(1, N = 258) = 0.25, p = 0.618; brand recall, χ²(1, N = 258) = 0.76, p = 0.385; brand recognition, χ²(1, N = 258) = 0.91, p = 0.341; having seen the video before, χ²(2, N = 258) = 5.25, p = 0.073; familiarity with the vlogger, χ²(2, N = 258) = 0.03, p = 0.984; video rating, t(256) = 0.76, p = 0.448, 95% CI [−0.24; 0.55], d = 0.01; vlogger rating, t(256) = 1.42, p = 0.157, 95% CI [−0.12; 0.74], d = 0.02; brand familiarity, t(256) = 0.41, p = 0.680, 95% CI [−0.23; 0.35], d = 0.01; and brand use, t(256) = 0.56, p = 0.579, 95% CI [−0.15; 0.27], d = 0.01. These variables were therefore not included as covariates in the main analyses. For the variable watching vlogs of this vlogger, we found a difference between the two conditions. In the experimental condition, children indicated that they watch vlogs of this vlogger less often (M = 1.55, SD = 0.80) than children in the control condition (M = 1.85, SD = 0.94), t(246.20) = 2.79, p = 0.006, 95% CI [0.09; 0.52], d = 0.03. However, this variable was not significantly correlated with brand attitude, r(256) = 0.06, p = 0.351, nor with any of the mediating variables (all p ≥ 0.113). Therefore, this variable was not included as a covariate in the subsequent main analyses.

Despite careful preparation, this study is subject to some limitations, leading to suggestions for future research. A first limitation is that the study implemented a single-message design. We decided to use a single-message design to keep the overall feasibility of the study high (i.e., chance of achieving required sample size) and the burden on the young participants acceptable. Moreover, by using a single-message design, we were better able to interpret any possible effects. However, a disadvantage of a single message design is that the effects, or lack of effects, could be driven by other message features (e.g., type of vlogger, gender of vlogger, narrative in the story). Future research could examine the hypothesized relationships for a variety of vlog messages. Doing so would allow to draw conclusions beyond one instance of a sponsored vlog.

Another limitation concerns the stimulus material, specifically the selected vlog and sponsoring brand, and the disclosure message used. In the current study, we used a vlog in which the brand was already quite prominent—this may have made the commercial intent of the vlog too obvious (Friestad and Wright, 1994; Van Reijmersdal, 2009). A suggestion for future research is to choose a vlog in which the brand and commercial intent is less prominent. If the commercial intent is less evident, then the effectiveness of the disclosure in increasing advertising literacy activation could be different because children actually need the disclosure to alert them to the commercial intent. This is also important because the commercial intent of in-vlog advertising is usually not as clearly presented as in the vlog used in the current study.

Another suggestion for future research related to the stimulus material is to work with a brand that is less familiar and less popular. In the current study, over 90% of the participants were familiar with the brand and the overall attitude toward the brand was very positive. This could be because children already had a favorable attitude toward the brand. As a consequence, it may have been difficult for the vlog and the disclosure to affect this existing attitude. The attitude toward a less familiar or unfamiliar brand may be easier to influence with exposure to a disclosure. Using a new and unfamiliar brand could therefore provide further insight into the relation between children’s dispositional advertising literacy, advertising literacy activation, advertising susceptibility, and the usefulness of a disclosure.

In future research, it is also important to focus on children’s understanding of the meaning of the disclosure. The present study used an explicit disclosure based on current regulations in the United States and Europe (e.g., FTC, 2013; WOMMA, 2013). The disclosure explicitly mentioned the name of the vlogger who created the video, the brand, the relationship between the two, and the fact that the brand paid for the advertising. However, it remains unknown if and how children understood the meaning of the disclosure. Differences in understanding of the disclosure may determine its effectiveness in activating persuasion knowledge. Future research could measure children’s understanding of the sponsorship disclosure used and include it as a control variable in the analyses. Furthermore, more research is needed into how disclosures should be designed and formulated to be understood by young viewers.

Another suggestion for future research with regard to the stimulus material used is to add a condition in which participants are shown a neutral ‘dummy’ message (e.g., “You are going to see an online video now”) at the start of the video. In doing so, a condition is added in which the structure of the message (i.e., a sentence presented on a black screen before the vlog content + vlog content) is exactly the same. As a result, several alternative explanations for a possible disclosure effect (e.g., the potential effect could be driven by the mere presence of a text before the video, not so much by the content of the text, and reading a text requires mental resources for processing which might consecutively affect children’s responses to the video) can be excluded.

With regard to the dependent measures for advertising literacy, future research could focus on other dimensions, representing more sophisticated levels of understanding (e.g., understanding of persuasive tactics used, understanding of the economic model of advertising) as well. In the current study, we chose to focus on two dimensions of conceptual ad literacy (i.e., advertising recognition and understanding of advertising’s selling and persuasive intent) and two dimensions of attitudinal advertising literacy (i.e., skepticism and critical attitude/disliking). The reason for this was that in the literature these variables are assumed to be important basic dimensions of advertising literacy, together forming the fundament of children’s advertising schemas (Rozendaal et al., 2011b; Hudders et al., 2017). Research shows that around the age of 7 (the minimum age of the children in the current study), most children have developed these fundamental elements of advertising literacy (De Jans et al., 2019b). For the current study, this was important because in order to activate advertising literacy from memory, children need to have a sufficient level of this literacy in place. The more sophisticated dimensions of advertising literacy, such as the understanding of persuasive tactics, develop at a significantly later age (Rozendaal et al., 2011a). Chances are that the youngest children in our sample had not yet developed this understanding, which means that they cannot activate it (regardless of whether they would actually do so if they were exposed to advertising).

Finally, there are some suggestions for future research with regard to the measurement methods used. In the current study two different measurement methods were used. Although it seems that the questionnaire measurement and indirect measurement yielded similar results, more research is needed on this topic. First, it is important to get a broader understanding whether the indirect and direct measure actually assess the exact same construct because this is not always clear (Fazio and Olson, 2003). It could be that the self-reported questionnaire measures a different level of advertising processing than the indirect measurement task. For instance, systematic processing may be better assessed with a direct measurement method while heuristic and automatic processing is better assessed with indirect measurement methods (Buijzen et al., 2010). Future research needs to test this.

Second, this work used a questionnaire measurement that provided support for the role of attitudinal advertising literacy activation in making children less susceptible to advertising effects whereas the indirect measurement did not. This finding could be further explored in future research that assesses brand attitude with both a questionnaire (as was done in this study) as well as with an indirect measurement task (e.g., an IAT or Approach-Avoidance Task). This will provide a better understanding of the relationship between different measurement methods in assessing children’s advertising literacy activation and their brand responses. For future research it is important to consider that the indirect measurement has several advantages over the questionnaire measurement—it is more difficult to give social desirable answers when an indirect measure is used (see Hoek et al., 2019 for an overview of advantages of an indirect measurement method as compared to a questionnaire measurement). However, indirect measurements also have their disadvantages. For instance, an indirect measurement task is more time-consuming and requires more technical skills from the researchers as compared to a questionnaire. Furthermore, it is paramount that the indirect measurement task is understood by the children as it is intended by the researchers.

The results of this study have implications for the scientific community as well as government guidelines and advertisers. First, this study showed that a textual disclosure reading “[name vlogger] is being paid by [name company] to advertise in this video” does not necessarily increase the extent to which children activate their advertising literacy when exposed to in-vlog advertising. In this study, this may be due to the fact that the brand is very prominent in the video. Prominently placed brands seem to be an inherent trigger for children to activate their advertising literacy regardless of the presence of a disclosure. However, whether the prominence of the brand indeed made it easier for the children to activate their advertising literacy remains to be tested in future research. An important scientific implication of this study therefore is that the effectiveness of a disclosure for in-vlog advertising should always be considered in relation to the specific features of the brand placement and the vlog characteristics (e.g., prominence of the brand, familiarity of the brand and the vlogger, prior attitude to the brand and the vlogger). These features can make it either easier or harder for children to use their retrieval strategies in order to activate their advertising literacy schemas. From a theoretical perspective it is therefore important to take the characteristics of the message into account when studying children’s advertising information-processing and schema activation abilities.

Another theoretical implication is that this study showed that both direct and indirect measures are needed to get a comprehensive view of the relation between children’s advertising literacy and advertising susceptibility. The present study showed that direct and indirect measurements of advertising literacy activation reveal different processes through which children make sense of, and are affected by, advertising. That is, direct measures reveal more conscious and deliberate ways of advertising processing, while indirect measures reveal more subconscious and automatic ways of advertising processing. In daily life, children only use few mental resources to process advertising messages, meaning that children’s advertising processing is rather automatic (Buijzen et al., 2010; Rozendaal et al., 2011b). Therefore, direct measurement tools may not be capable of providing an answer to the question how children process advertising in a ‘natural’ state (e.g., when they are at home and not in an experimental setting, filling out a questionnaire). Indirect measures may be better capable to reveal this natural and more automatic way of advertising processing. Therefore, an important scientific implication is that to obtain full understanding of the role of advertising literacy in children’s susceptibility to advertising effects, both direct and indirect measures are needed.

Furthermore, our findings have important practical implications because such disclosures are typically required in current advertising guidelines in Europe and the United States. The current study showed that the disclosure, as used here, has failed to further raise the level at which children activate their advertising literacy while watching a vlog. This is not to say that all sponsorship disclosures are ineffective. It is important to first get an understanding under what circumstances sponsorship disclosures are effective in activating children’s advertising literacy. Based on these insights, government agencies might reconsider their guidelines and investigate whether other types of disclosures are more effective in increasing children’s advertising literacy activation, especially when the in-vlog advertising is relatively subtle and less prominent.