Inner Speech and Grit: Do Positive Inner Speech and Evaluative Inner Speech Lead to Grit Behaviour
Dublin Core
Title
Inner Speech and Grit: Do Positive Inner Speech and Evaluative Inner Speech Lead to Grit Behaviour
Creator
Huzaifah Adam
Date
2023
Description
Grit, defined as perseverance and passion for long-term goals, is a reliable predictor of success metrics, surpassing even IQ. While the exploration of grit has been conducted extensively, studies on the mechanisms of grit are still lacking. Inner speech, the silent production of words in one’s mind, plays a pivotal role in managing thoughts. This includes cognitive reframing, which is essential for enhancing perseverance. Theoretically, inner speech can predict grit. This study, employing a survey and experimental design, aims to investigate whether positive inner speech and evaluative inner speech can predict grit behaviour. The data for this study (n=56) were collected in two ways: (1) using the grit scale and inner speech VISQ-R via a Qualtrics survey, and (2) using participants’ task retention decisions and a qualitative classification approach. The data were analysed using R Studio. The survey data were analysed via a linear model, while the qualitative data were analysed using a generalised linear mixed-effects model. The survey results showed that only evaluative inner speech can positively predict grit. However, there were imbalanced results regarding the participants’ task retention decisions. Collectively, these findings underscore that grit can be predicted by evaluative inner speech. This prompts further research to explore its multifaceted role in shaping grit across various domains.
Subject
Inner speech, grit, articulatory suppression
Source
This study applied a mixed-method and correlational research design that aims to examine whether evaluative inner speech and positive inner speech lead to grit behaviour. The data for this study were collected using two methods: (1) questionnaires through a Qualtrics survey, and (2) an experimental task where the participants were asked to complete two sets of puzzles under different conditions (baseline and with articulatory suppression) and provide their retrospective experience after each puzzle task. Participants’ task retention decisions (decision to quit) were also recorded in the study. Three different analyses were applied in the research. For the first analysis, the positive inner speech and evaluative inner speech scores from VISQ-R acted as the predictors, and grit from the Short Grit Scale as the outcome. For the second analysis, the participant’s grit score acted as the predictor, and the participant’s task retention decision acted as the outcome. Lastly, the third analyses the types of inner speech based on the participant’s retrospective experience (positive inner speech and evaluative inner speech) acted as the predictors, and the participant’s decision to quit or not to quit was the output.
In this study, the participants were students from Lancaster University, ranging from undergraduate degree students to master’s degree students and doctorate students. Participants were recruited using social networks, direct emails, and posters around the campus and/or on social media. The session took approximately 30 minutes for the data collection process, including the briefing, and each participant was reimbursed with five GBP for participating. Ethical approval for this study was submitted and approved by the ethics committees at Lancaster University.
The number of participants involved in the study was 56 people in total. This number was determined by using G power. The test family was set at the t-test because this research will use a comparison between the control approach (baseline) and the experimental approach (with articulatory suppression). The effect size f2 was set at 0.15, while the α-error probability was set to 0.05 (5%) and the power 1−β error of probability at 0.8 (80%), with the number of predictors set at five. In total, 56 participants took part in the study, where the number of male and female participants was 23 (41%) and 33 (59%), respectively, and the number of native English participants in the study was 15 (27%), while non-native speakers were 41(73%).
Demographic Information: The demographic information collected pertained to each
participant’s attributes. This included sex (male, female, non-binary/third gender, and prefer not to say) and English native background (yes or no). Although the study has no biases towards the participant’s native language, the word used in the study ‘aluminium’, a word that is suggested by Gathercole and Baddeley (2014) for the research, may or may not influence the fluidity of pronunciation, making the articulatory suppression more challenging for non-native speakers.
Varieties of Inner Speech Questionnaires Revised (VISQ-R): The VISQ-R was developed to link the everyday phenomenology of inner speech, including any psychopathological traits and inner dialogue (Alderson-Day et al., 2018). There are two versions of the Varieties of Inner Speech Questionnaire, where the original one consisted of 18 items and the revised version VISQ-R consisted of 26 items (see Appendix D) that took approximately 5-8 minutes to be completed via a Qualtrics survey. In this study, VISQ-R has been presented as internal experience questions as a dummy to the real name. This is to eliminate any possible biases by the respondents.
Responses from VISQ-R can be subdivided into five dimensions and into seven scales (Not like me at all – Very much like me) for scoring: dialogical, evaluative, condensation, other people, and positive. A higher score in dialogical indicates that the person often uses inner speech to exchange ideas with oneself and vice-versa. A higher evaluative score means that the person often uses inner speech to evaluate their thoughts, actions, and decisions. For condensation, a higher score indicates that a person talks to themselves in a concise or short words manner to encapsulate complex thoughts or ideas. Meanwhile, a higher ‘other people’ score indicates that a person often imagines other people’s voices or opinions when engaging in inner speech. Lastly, a high positive score indicates that the person often uses inner speech to encourage oneself in a supportive and comforting manner. Subscale totals for each dimension were acquired by adding the scores for each subscale and dividing it by the total number of items answered across the respective subscale.
The Varieties of Inner Speech Questionnaire has been supported for its reliability and validity in measuring inner speech. Racy et al. (2022) have studied the reliability of VISQ-R and compared it to six other instruments relating to inner speech. VISQ-R has moderate to strong concurrent validity with other instruments with self-evaluation showing a strong correlation with other measures. The internal consistencies and reliabilities were excellent (Cronbach’s α > .80) for each of the dimensions with only a positive dimension that is slightly lower with moderate to high test-retest reliability (>.60) (Alderson-Day et al., 2018).
Short Grit Scale (Grit-S): The questionnaire of Grit-S was developed by Angela Duckworth to measure the trait level of perseverance and passion for long-term goals (Duckworth & Quinn, 2009). The Grit-S consisted of eight items of questions (See Appendix D) with four fewer items in comparison to the original version, retaining the factor structure and improving on the psychometric properties. The questionnaire needs an approximation of 3- 5 minutes to be completed in the Qualtrics survey. Similar to VISQ-R, the Grit-S questionnaire has been presented as a personality instead of a grit scale to avoid any possible biases.
There are two dimensions included in the Grit-S for scoring: Consistency of Interest, where a higher scale subscale score indicates that the individual is able to maintain their interest for and focus on their long-term goal, and Perseverance of Effort, where a higher subscale score represents sustained effort towards a long-term goal despite the presence of setbacks (Van Doren et al., 2019). The subscale for the dimension of Consistency of Interest is acquired by adding the scores for all the subscale items (item-1, item-3, item-5, and item-6), while for Perseverance of Effort (item-2, item-4, item-7, and item-8). There are a few items that have been coded inversely and have been recoded before running the analysis.
Several research studies have confirmed the validity and reliability of the Short-Grit Scale Instrument. Eskreis-Winkler et al. (2014) conducted a study involving predicting retention in the military where the grit instrument was used to measure the grit level of cadets. The instrument has been proven to be reliable as grittier soldiers were more likely to complete the Army Special Operation Forces (ARSOF), likely to get a job, and likely to stay married. In a more recent study by Priyohadi et al. (2019), the Grit-S again proved its validity and consistency. The internal consistencies between items in a dimension were moderate to high (>.60) for both persistence of effort and consistency of interest and have high consistencies between studies.
Active Task: The jigsaw puzzle was used as the active task for this research. Two jigsaw puzzles from Livewire Puzzles were predetermined by the website as expert-level with 70 puzzle pieces (10 X 7) with an 8-minute time limitation. The puzzle can be accessed through the games.puzzle.ca website. The puzzles have been created by Arkadium, a company that is well-recognised in making online games. New puzzles have been uploaded daily, but to avoid any possible advantage or disadvantage, the puzzles used are from the 22nd of June 2023 and 21st of June 2023. Marks will also be provided at the end of each puzzle.
There are two ways of measuring participants’ performance: (1) Quitting - participants were allowed to quit the task at any time during the 8-minute time limit by telling the researcher present that they want to stop, and (2) Puzzle performance - marks will be given at the end of the puzzle (marks will be given even if participants quit halfway) by the source website. The marks will be calculated based on the number of puzzles fixed correctly and then divided by the total number of unfixed puzzles and will be multiplied by the amount of time left in the puzzle. The maximum score of the puzzle is 5,000 and the minimum score is zero. All calculations will be automatically measured by the source website.
The puzzle from Livewire Puzzle has also been used by other studies that focus on measuring grit using an active task. Kalia et al. (2019), similar to this study, used puzzles from Livewire Puzzle as an active task to measure perseverance in participants. Instead of using a jigsaw puzzle, Kalia opted to use sudoku to measure the role of grit and cognitive flexibility 2.4 Procedure
The research took place in one-on-one sessions at the Lancaster University library. Data collection sessions were administered in the following order: demographic information, the first puzzle task, the difficulty level question, the subjective inner speech question, the second puzzle task, the second puzzle difficulty level question, and finally, the second subjective inner speech question. Each participant undertook the puzzle task in both control (baseline) and experimental conditions (with articulatory suppression). The sequence of which puzzle task they had to complete first was decided based on the participant’s subject ID assigned by the researcher. Participants with odd Subject ID numbers were assigned the control puzzle task first, while participants with even Subject ID numbers were assigned the experimental puzzle task first. Before starting the experimental puzzle task, the researcher spent a few minutes helping the participants practice performing the articulatory suppression by saying the word ‘aluminium’ repeatedly at 90 BPM using an online metronome. Throughout the experimental task, if the participants mispronounced the word too obviously or consistently missed or skipped a beat, the researcher aided them by correcting their pronunciation or assisting them to meet the 90 BPM until they matched the rhythm again.
During data collection, the researcher offered participants an opportunity for a break between puzzles if they began to get tired to prevent their answers from being expedited. The participants were also allowed to ask any questions while they were completing the questionnaire to clarify their understanding of the items presented. At the end of each data collection session, the researcher thanked the participants for their participation and answered any questions that they had. The researcher also explained that participants would be emailed a participant debrief sheet and could request a summary of the study’s findings once data analyses had been completed. For participants who were eligible for reimbursement of travel expenses, they were asked to fill out a participant payment form as a receipt of confirmation that they had been paid.
Three different models of analysis were carried out in the study. To measure the first prediction, a linear model was used by entering the positive inner speech and evaluative inner speech scores from the VISQ as the predictors and the grit score from the short grit scale as the output. For the second prediction, a linear model was used with the outcome set at the participant’s decision to quit or not to quit and the predictor set as the interaction between different experimental conditions and grit. To measure the third prediction, a generalised linear mixed-effect model was explored by entering the interaction of different experimental conditions and dimensions of inner speech (evaluative inner speech and positive inner speech) recorded from the participant’s retrospective experience as the predictor and participant’s decision to quit the task as the outcome. In this model, a random effect of differences between the conditions (baseline and with articulatory suppression) in slope and participants in the intercept were also included.
In this study, the participants were students from Lancaster University, ranging from undergraduate degree students to master’s degree students and doctorate students. Participants were recruited using social networks, direct emails, and posters around the campus and/or on social media. The session took approximately 30 minutes for the data collection process, including the briefing, and each participant was reimbursed with five GBP for participating. Ethical approval for this study was submitted and approved by the ethics committees at Lancaster University.
The number of participants involved in the study was 56 people in total. This number was determined by using G power. The test family was set at the t-test because this research will use a comparison between the control approach (baseline) and the experimental approach (with articulatory suppression). The effect size f2 was set at 0.15, while the α-error probability was set to 0.05 (5%) and the power 1−β error of probability at 0.8 (80%), with the number of predictors set at five. In total, 56 participants took part in the study, where the number of male and female participants was 23 (41%) and 33 (59%), respectively, and the number of native English participants in the study was 15 (27%), while non-native speakers were 41(73%).
Demographic Information: The demographic information collected pertained to each
participant’s attributes. This included sex (male, female, non-binary/third gender, and prefer not to say) and English native background (yes or no). Although the study has no biases towards the participant’s native language, the word used in the study ‘aluminium’, a word that is suggested by Gathercole and Baddeley (2014) for the research, may or may not influence the fluidity of pronunciation, making the articulatory suppression more challenging for non-native speakers.
Varieties of Inner Speech Questionnaires Revised (VISQ-R): The VISQ-R was developed to link the everyday phenomenology of inner speech, including any psychopathological traits and inner dialogue (Alderson-Day et al., 2018). There are two versions of the Varieties of Inner Speech Questionnaire, where the original one consisted of 18 items and the revised version VISQ-R consisted of 26 items (see Appendix D) that took approximately 5-8 minutes to be completed via a Qualtrics survey. In this study, VISQ-R has been presented as internal experience questions as a dummy to the real name. This is to eliminate any possible biases by the respondents.
Responses from VISQ-R can be subdivided into five dimensions and into seven scales (Not like me at all – Very much like me) for scoring: dialogical, evaluative, condensation, other people, and positive. A higher score in dialogical indicates that the person often uses inner speech to exchange ideas with oneself and vice-versa. A higher evaluative score means that the person often uses inner speech to evaluate their thoughts, actions, and decisions. For condensation, a higher score indicates that a person talks to themselves in a concise or short words manner to encapsulate complex thoughts or ideas. Meanwhile, a higher ‘other people’ score indicates that a person often imagines other people’s voices or opinions when engaging in inner speech. Lastly, a high positive score indicates that the person often uses inner speech to encourage oneself in a supportive and comforting manner. Subscale totals for each dimension were acquired by adding the scores for each subscale and dividing it by the total number of items answered across the respective subscale.
The Varieties of Inner Speech Questionnaire has been supported for its reliability and validity in measuring inner speech. Racy et al. (2022) have studied the reliability of VISQ-R and compared it to six other instruments relating to inner speech. VISQ-R has moderate to strong concurrent validity with other instruments with self-evaluation showing a strong correlation with other measures. The internal consistencies and reliabilities were excellent (Cronbach’s α > .80) for each of the dimensions with only a positive dimension that is slightly lower with moderate to high test-retest reliability (>.60) (Alderson-Day et al., 2018).
Short Grit Scale (Grit-S): The questionnaire of Grit-S was developed by Angela Duckworth to measure the trait level of perseverance and passion for long-term goals (Duckworth & Quinn, 2009). The Grit-S consisted of eight items of questions (See Appendix D) with four fewer items in comparison to the original version, retaining the factor structure and improving on the psychometric properties. The questionnaire needs an approximation of 3- 5 minutes to be completed in the Qualtrics survey. Similar to VISQ-R, the Grit-S questionnaire has been presented as a personality instead of a grit scale to avoid any possible biases.
There are two dimensions included in the Grit-S for scoring: Consistency of Interest, where a higher scale subscale score indicates that the individual is able to maintain their interest for and focus on their long-term goal, and Perseverance of Effort, where a higher subscale score represents sustained effort towards a long-term goal despite the presence of setbacks (Van Doren et al., 2019). The subscale for the dimension of Consistency of Interest is acquired by adding the scores for all the subscale items (item-1, item-3, item-5, and item-6), while for Perseverance of Effort (item-2, item-4, item-7, and item-8). There are a few items that have been coded inversely and have been recoded before running the analysis.
Several research studies have confirmed the validity and reliability of the Short-Grit Scale Instrument. Eskreis-Winkler et al. (2014) conducted a study involving predicting retention in the military where the grit instrument was used to measure the grit level of cadets. The instrument has been proven to be reliable as grittier soldiers were more likely to complete the Army Special Operation Forces (ARSOF), likely to get a job, and likely to stay married. In a more recent study by Priyohadi et al. (2019), the Grit-S again proved its validity and consistency. The internal consistencies between items in a dimension were moderate to high (>.60) for both persistence of effort and consistency of interest and have high consistencies between studies.
Active Task: The jigsaw puzzle was used as the active task for this research. Two jigsaw puzzles from Livewire Puzzles were predetermined by the website as expert-level with 70 puzzle pieces (10 X 7) with an 8-minute time limitation. The puzzle can be accessed through the games.puzzle.ca website. The puzzles have been created by Arkadium, a company that is well-recognised in making online games. New puzzles have been uploaded daily, but to avoid any possible advantage or disadvantage, the puzzles used are from the 22nd of June 2023 and 21st of June 2023. Marks will also be provided at the end of each puzzle.
There are two ways of measuring participants’ performance: (1) Quitting - participants were allowed to quit the task at any time during the 8-minute time limit by telling the researcher present that they want to stop, and (2) Puzzle performance - marks will be given at the end of the puzzle (marks will be given even if participants quit halfway) by the source website. The marks will be calculated based on the number of puzzles fixed correctly and then divided by the total number of unfixed puzzles and will be multiplied by the amount of time left in the puzzle. The maximum score of the puzzle is 5,000 and the minimum score is zero. All calculations will be automatically measured by the source website.
The puzzle from Livewire Puzzle has also been used by other studies that focus on measuring grit using an active task. Kalia et al. (2019), similar to this study, used puzzles from Livewire Puzzle as an active task to measure perseverance in participants. Instead of using a jigsaw puzzle, Kalia opted to use sudoku to measure the role of grit and cognitive flexibility 2.4 Procedure
The research took place in one-on-one sessions at the Lancaster University library. Data collection sessions were administered in the following order: demographic information, the first puzzle task, the difficulty level question, the subjective inner speech question, the second puzzle task, the second puzzle difficulty level question, and finally, the second subjective inner speech question. Each participant undertook the puzzle task in both control (baseline) and experimental conditions (with articulatory suppression). The sequence of which puzzle task they had to complete first was decided based on the participant’s subject ID assigned by the researcher. Participants with odd Subject ID numbers were assigned the control puzzle task first, while participants with even Subject ID numbers were assigned the experimental puzzle task first. Before starting the experimental puzzle task, the researcher spent a few minutes helping the participants practice performing the articulatory suppression by saying the word ‘aluminium’ repeatedly at 90 BPM using an online metronome. Throughout the experimental task, if the participants mispronounced the word too obviously or consistently missed or skipped a beat, the researcher aided them by correcting their pronunciation or assisting them to meet the 90 BPM until they matched the rhythm again.
During data collection, the researcher offered participants an opportunity for a break between puzzles if they began to get tired to prevent their answers from being expedited. The participants were also allowed to ask any questions while they were completing the questionnaire to clarify their understanding of the items presented. At the end of each data collection session, the researcher thanked the participants for their participation and answered any questions that they had. The researcher also explained that participants would be emailed a participant debrief sheet and could request a summary of the study’s findings once data analyses had been completed. For participants who were eligible for reimbursement of travel expenses, they were asked to fill out a participant payment form as a receipt of confirmation that they had been paid.
Three different models of analysis were carried out in the study. To measure the first prediction, a linear model was used by entering the positive inner speech and evaluative inner speech scores from the VISQ as the predictors and the grit score from the short grit scale as the output. For the second prediction, a linear model was used with the outcome set at the participant’s decision to quit or not to quit and the predictor set as the interaction between different experimental conditions and grit. To measure the third prediction, a generalised linear mixed-effect model was explored by entering the interaction of different experimental conditions and dimensions of inner speech (evaluative inner speech and positive inner speech) recorded from the participant’s retrospective experience as the predictor and participant’s decision to quit the task as the outcome. In this model, a random effect of differences between the conditions (baseline and with articulatory suppression) in slope and participants in the intercept were also included.
Publisher
Lancaster University
Format
The data format is csv.
Identifier
Adam2023
Contributor
Huzaifah Adam
Rights
Open
Language
English
Type
Data
Coverage
LA1 4YF
LUSTRE
Supervisor
Dr. Bo Yao
Project Level
MSc
Topic
Developmental
Sample Size
56 Participants
Statistical Analysis Type
Linear Model, Qualitative
Files
Collection
Citation
Huzaifah Adam, “Inner Speech and Grit: Do Positive Inner Speech and Evaluative Inner Speech Lead to Grit Behaviour,” LUSTRE, accessed April 28, 2024, https://www.johnntowse.com/LUSTRE/items/show/180.