SHAREOK Repository
Not a member yet
49261 research outputs found
Sort by
Misinformation Detection Methods Using Large Language Models and Evaluation of Application Programming Interfaces
Get PDFMisinformation has emerged as a pressing public policy concern, prompting transdisciplinary research in the data science field. News journalism provides a foundation for free speech in modern society, yet misinformation in mainstream and independent media through opinionated or biased news can pose dangerous consequences ranging from misunderstanding basic facts to emboldened extremism. Currently, the preeminent tool for misinformation detection is the large language model (LLM) as it is renowned for its ability to capture the context and meaning of textual data. In addition, generative artificial intelligence (AI) models, namely OpenAI's ChatGPT and Google's Pathways Language Model (PaLM), are accessible in an application programming interface (API) form, which can also provide opportunities for automated misinformation detection. Despite advancements in developing effective data science tools for identifying misinformation, there are not many available options, and it is crucial to assess pre-existing tools to determine the most recommended model to pursue for future field research and open-source use in automated misinformation detection. This thesis attempts to evaluate fine-tuned supervised LLMs, AI model frameworks, and unsupervised learning methods to propose an explainable, automated misinformation detection tool that incorporates multiple natural language processing (NLP) dimensions and holistically evaluates trustworthiness in news articles. The study revealed that the Hugging Face LLM RoBERTa with added NLP dimensions as features was the most effective model. Furthermore, it was found that unsupervised learning methods provided valuable insights that eliminated some ambiguity between trustworthy and fake news articles, and AI models tended to inflate the trustworthiness values of news articles.
Keywords: misinformation, large language models (LLMs), unsupervised learning, application programming interfaces (APIs
TICK ABUNDANCE AND HABITAT DYNAMICS IN URBAN PARKS: A STUDY OF MICRO- AND MACRO- HABITAT INFLUENCES IN THE OKLAHOMA CITY METROPOLITAN AREA
Get PDFMost studies on tickborne diseases in the U.S. have been concentrated in suburban and rural areas. This has led to a lack of understanding about the risks of such diseases in urban settings, where environmental conditions vary, and human-tick interactions differ significantly. Several previous studies, including a recent study in Oklahoma City in 2017-2018, have investigated how microhabitat conditions within sampling locations can impact tick abundance. However, microhabitat conditions only capture a limited spatiotemporal range, typically confined to the immediate sampling site and period. Monitoring macro-environmental conditions via Earth observations could enhance our understanding of tick ecology on a larger scale. Here, we hypothesize that integrated monitoring of micro- and macro-habitat conditions can better capture tick abundance in urban parks. Specifically, we hypothesize that tick abundance in urban parks is positively influenced by microclimate factors like higher humidity levels or deeper leaf litter, and by macro-environmental conditions such as habitat type and spatial arrangement. We collected ticks from 13 parks in the Oklahoma City Metropolitan area using CO2 traps and flagging techniques. At sampling transects, we gathered micro-environmental data, including temperature, humidity, wind speed, and vegetation types, and assessed macro-environmental conditions such as land cover, incorporating landscape metrics. Our analysis identified key factors influencing tick abundance in urban parks, revealing that solar radiation negatively affects tick abundance, while soil pH, woody vegetation, and forest cover have positive effects. For adult ticks, leaf litter depth and solar radiation are negative predictors, while woody vegetation and forest percentage positively influence their abundance. Nymph tick abundance increases with lower solar radiation and relative humidity and is positively correlated with soil pH. Overall, both micro-environmental variables and landscape metrics are crucial in predicting tick abundance across different life stages. The results underscore the complex interplay between environmental factors and tick distribution, with implications for public health strategies in urban planning and park management
THE IMPACT OF REMOTE AND LOCALIZED KNEE PAIN ON MAXIMAL STRENGTH, FATIGUE PARAMETERS, AND FORCE RECOVERY IN THE KNEE EXTENSOR MUSCLES OF RECREATIONALLY ACTIVE COLLEGE AGED INDIVIDUALS
Get PDFPain is a widely experienced sensation that has been shown in previous research to decrease duration of exercise and maximal strength. PURPOSE: The purpose of this study was to determine if the application of knee pain impacts time-to-task failure and maximal strength measures. It also aimed to see if changing the location of knee pain between the ipsilateral (dominant) leg and the contralateral (non-dominant) impacted fatigue parameters, maximal strength differently and force recovery. METHODS: Participants (Females = 9, Males = 12, N/A = 1) completed five visits (2 familiarization and 3 experimental) separated by 48 hours. At the beginning of each of the visits, the electrical stimulus for participants to reach pain threshold and pain rated a 4/10 in the knee was obtained. The exercise protocol consisted of four pre-exercise MVCs (Maximal Voluntary Contractions), submaximal time-to-task failure protocol (intensity of 15% over critical torque), and 6-post exercise MVCs with the different conditions being contralateral knee pain, ipsilateral knee pain or no knee pain. RESULTS: The male participants had a higher MVC and twitch torque than the female participants (p<0.001) indicating males were stronger than women. Pre-exercise MVCs decreased significantly following application of knee pain, likely due to an increase in central fatigue (p<0.001). There was a main effect of time in force decrease (p<0.001), twitch torque (p<0.001), and % activation (p=0.005) during submaximal protocol. Regardless of pain condition, force output was fully recovered 240 seconds after exercise failure (p=0.067). Subjective pain measures of exercising muscle pain, regardless of condition, were increased at 60 seconds (p=0.030), immediately before failure (p=0.004), and immediately following exercise (p=0.007). When pain was applied to the ipsilateral knee, pain ratings decreased significantly over time, likely due to exercise induced hypoalgesia (p<0.001). Pain ratings in the ipsilateral knee were greater in the ipsilateral condition than they were in the control condition or the contralateral knee pain condition (p<0.001). When pain was applied to the contralateral knee, pain ratings were different over time also likely due to exercise induced hypoalgesia (p<0.001). Pain ratings in the contralateral knee were greater in the contralateral condition than they were in the control condition or the ipsilateral knee pain condition (p<0.001). CONCLUSIONS: The presence of knee pain regardless of location decreased MVCs and increased percent activation of the participants. Knee pain stimulus regardless of location did not impact time-to-task failure. Peripheral fatigue increased throughout the fatiguing exercise protocol. Maximal strength was recovered 240s after fatiguing exercise regardless of the presence and location of knee pain
An Investigation of Organizational Citizenship Behavior De-escalation
No full textAlthough prior research indicates that employees often feel pressure to increase levels of organizational citizenship behavior (OCB), very little systematic research has explored the opposite occurrence – when employees decrease levels of OCB. Thus, in this dissertation, I use qualitative and quantitative methodology to investigate the lived experiences of employees who engage in OCB de-escalation. Through thematic analysis, I develop a process model to explain why and how employees reduce OCB, which specific behaviors are reduced, and what consequences they experience as a result. I advance theory on OCB by highlighting the ways that fluctuations in OCB influence employees’ experiences at work. My findings also hold practical relevance, as organizations seek to curb recent trends of decreased engagement and to encourage good citizenship at work
Coevolution of a male novel genital trait and female reproductive response in Drosophila melanogaster sister species
No full textNovel morphological structures have always fascinated scientists. Developmental processes give rise to structures and morphologies that are often functionally necessary for an organism’s survival and reproduction. But how does evolution shape these developmental processes to give rise to new phenotypes? How do these new phenotypes ultimately give rise to new species? And, how do these new phenotypes interact and coevolve between the sexes, particularly in the context of reproduction? Most studies that have focused on the evolution of novel traits focus on the importance of transcription factors and their pathways, and many studies have shown that evolution in regulatory regions and/or coding regions of transcription factors often gives rise to differences between species. In this dissertation, I investigated how novel traits diverge among the four sister species of Drosophila melanogaster and how novel traits are shaped by female sensory variation, focusing on an unconventional candidate.
In Chapter 1, I used a genetic introgression line called 4C2(A) to identify genes that are involved in the development of a novel male reproductive structure of the external genitalia of the four sister species. The introgression lines were created by crossing a D. mauritiana male carrying a P-element with a D. sechellia white (w) female and back-crossing it to D. sechellia w male for 15 generations to have a small D. mauritiana region in an otherwise D. sechellia w genetic background. These secondary sexual structures are called epandarial posterior lobes (ePLs) and help the male to grasp the female during mating and ensure a tight genital coupling. Using a series of mapping experiments and functional genetic tests, I identified Allnighter (Aln), a pseudokinase, as a gene that directs the development of ePLs. Using additional functional genetic experiments, I showed that Aln is necessary for specifying ePL size differences among the four sister species. Furthermore, our data showed that Aln had a more prominent role when flies developed under stressful thermal conditions. Specifically, when Aln null individuals developed in higher temperatures, they exhibited an even smaller ePL size and an incomplete genitalia rotation, resulting in upside-down male genitalia. This work suggests the co-option of the cellular stress response pathways can result in divergent morphologies and highlights the role of cellular stress response pathways in evolutionary developmental biology.
Most, if not all, traits on the male genitalia are under sexual selection, and no study will be complete without investigating how females respond to and co-evolve with these novel male reproductive traits. Previous studies have shown that when D. melanogaster females mate with males that have smaller or misshapen ePLs, they reduce the number of eggs they lay. In Chapter 2, I studied whether the gene or genes that specify ePLs size and female egg-laying are in the same region or even the same gene. I investigated the effect of the entire collection of available D. sechellia-D. mauritiana introgression lines that had smaller or misshapen ePLs on female egg-laying, with a focus on 4C2(A). We showed that when 4C2(A) females mate with 4C2(A) males, they do not reduce their egg-laying amounts, which suggests that a gene or genes in this region affect the female egg-laying phenotype. I tested the egg-laying amount of Aln null females, and our results show that these females laid significantly fewer eggs compared to the control females. These results demonstrate that Aln specifies not only male genital morphology, but also female reproductive behavior, providing a basis for sexual selection to act on males and females in concordance. These findings have important implications for our understanding of the co-evolution of male and female reproductive traits.
In Chapter 3, I delved deeper into the egg-laying phenotype and how females control this key reproductive event. I conducted a literature review and summarized the studies that show how specific neurons, whether from the peripheral nervous system or the central nervous system, modulate the action of egg deposition or the movement of the egg into the oviduct and other phenomena in the egg-laying behavior. Furthermore, by looking at other insect examples, I show that there are more components in the egg-laying behavior that require further study in Drosophila. Neuronal control of egg-laying in Drosophila provides many understudied avenues and has the potential to provide more robust evidence for phenomena like cryptic female choice.
In Chapter 4, I further explored how females sense the ePLs during mating and how they regulate their reproductive output/responses as a consequence of variation in ePL morphology. First, I identified candidate sensory neurons on female genitalia at the site where the ePLs make contact with the female during copulation. The candidate neurons I identified are mechanosensory sensilla on the epigynial ventral lobe. Interestingly, these bristles vary in average number among the four sister species. I showed that the neurons attached to the sensilla express Piezo, a mechanoreceptor gene, and are cholinergic. My results showed that Piezo knockout females mate normally but lay significantly fewer eggs compared to control females. On the other hand, although Piezo knockout males sire slightly fewer offspring, the difference in the number of eggs laid by females mated to Piezo knockout males compared to females mated to control males is not significant. Laser ablation of only some of these epigynial sensilla does not affect the female egg-laying output. However, when all of these sensilla were ablated, females laid close to no eggs. Additionally, we showed that axons of these neurons arborize the abdominal neuromere through the main abdominal nerve. The abdominal neuromere houses both motor neurons that modulate the female egg-laying and interneurons that carry the sensory signal to the brain, suggesting that the information carried by the epigynial sensilla neurons has the potential to influence the output of central nervous system neurons. Taken together, these neurons are promising candidates for future studies about sensing ePLs. In summary, my studies showed that the co-evolutionary dynamics between male and female reproductive traits are complex. Variation in peripheral sensory neurons can drive the evolution and diversification of novel traits and change in a gene like Aln not only provides the developmental raw material to diversify the size of the ePLs in the males but also reinforces the egg-laying bias in females
PLANT-SOIL FEEDBACK WITH DROUGHT LEGACY SOIL PROMOTES RESOURCE ACQUISITION OVER CONSERVATIVE STRATEGY IN A GRASSLAND ECOSYSTEM
No full textThe relationships between plants and soil (plant-soil feedback, or PSF) are increasingly being investigated for their significant influence on vegetation diversity and ecosystem functioning. Although there is increasing understanding of PSF within plant functional groups and species origins, the effect of drought on PSF returns inconsistent results, at a community level. The Southern US Great Plains is a water-limited system and therefore sensitive to changes in precipitation regime. As drought conditions are expected to increase in the Great Plains and beyond, we intend to elucidate the complex mechanisms of PSF and their role in individual plant response to drought, as well as community strategy. Plant traits are often used to infer ecological strategy of individuals and communities, and can capture plant responses to PSF. The leaf economic spectrum indicates ecological strategies ranging from conservative (slow) to acquisitive (fast) resource use, and vary between functional groups and climatic conditions. Since PSF has the capacity to influence the rate of plant resource use, we hypothesized that drought legacy soil would inhibit plant growth and function due to a potential limitation of beneficial microbiological mutualisms under drought, and that community level strategy would shift toward resource conservatism, or slow resource use, for the same reason. We replicated two plant communities from the mixed-grass prairie in central Oklahoma: one which has been subject to long-term experimental drought for six years, and a reference plant community. We conditioned both plant communities in heterospecific and conspecific soil /(using soil cores collected from their respective drought treatment), and uninoculated soil. We measured individual plant response by biomass, growth rate, and leaf traits, and community response using community weighted means (CWM), weighting traits by biomass. At an individual level, we saw no difference in plant response to heterospecific and conspecific soils, but that plants generally
had better response to inoculated soil than uninoculated soil. We also found shifts in biomass aboveground in reference plant communities when conditioned with drought soil. At a community level, we found that drought soil shifted ecological response of reference plant communities from resource conservative (in conspecific soil) to resource acquisitive, as indicated by leaf trait CWMs. Counter to our second hypothesis, this result suggests that drought legacy promotes a faster resource use strategy to plants
Introduction to Process Simulation with Aspen Plus: Instructor's Guide
Get PDFProcess simulation entails the representation of a chemical process by using mathematical equations. Process simulation in Aspen Plus offers the advantages of optimizing plant performance, reducing operational costs, and improving product quality through precise modeling and analysis of chemical processes. This book was written as an educational material to help Chemical/ Environmental/Process Engineering students working on their capstone project to learn more about process design. Capstone courses are designed to help students understand process design, preliminary economic assessment and process plant optimization. The book delves into the fundamentals of process design and offers practical examples for easy understanding. This book was used with ENGR 3440, mentored research course on process design at the University of Oklahoma.This project was funded by the University of Oklahoma Libraries' Alternative Textbook Gran
INTERFACIAL STABILITY WITH SURFACE-ACTIVE AGENTS USING COARSE-GRAINED MODELING
No full textThe stability of interfaces in the presence of surface-active agents such as surfactants and nanoparticles (NPs) is imperative in various industrial and environmental contexts. These include separation processes, enhanced oil recovery, and environmental remediation, highlighting the significance of interface stability in diverse scenarios. In stabilizing oil-water interfaces, surfactants and NPs play a vital role, but with different mechanisms in each case. NPs affect interfacial behavior by reducing the interfacial area, whereas surfactants lead to a reduction in interfacial tension of the oil-water system. Although many studies have concentrated on equilibrium conditions without considering fluid flow, industrial applications typically involve displacing immiscible fluids. Therefore, there is a need to explore the impact of flow conditions on the properties of the oil-water interface, especially in the context of the presence of surfactants and NPs. In this thesis, we investigated the effect of anionic surfactant (sodium dodecylsulfate, SDS) and nonionic surfactant (octaethylene glycol monododecyl ether, C12E8) on the oil-water interface by using Dissipative Particle Dynamics (DPD) simulation methods. In particular, SDS represented the case of a short surfactant molecule on the interface, while C12E8 was a longer molecule. Surfactants are in conjunction with NPs (homogeneous and anisotropic particles) either on the flow of oil and water in a narrow channel under Poiseuille or Couette flow conditions.
It is important first to verify the integrity of the computations by comparing the velocity profile resulting from simulations of Poiseuille and Couette flows to theoretical expectations. Based on this approach, satisfying the no-slip boundary conditions required the following: (1) creating frozen solid walls, (2) applying the bounce-back boundary condition, and (3) determining the DPD parameters necessary for ensuring that the boundary conditions were obeyed, that the oil and water viscosities were represented correctly. Moreover, the effects of surfactants at the oil-water interface and the interfacial instability under shear were investigated. A critical shear rate was found for Poiseuille flow, beyond which the surfactants desorb from the interface, forming micelles and destabilizing the interface. The surfactant-covered interface remained stable under Couette flow even at high shear rates. Besides the flat interface of oil and water, the motion of an oil droplet with and without the presence of surfactants and Janus particles (JPs) under shear was studied.
In certain cases, oil droplets are observed to be mobilized along a solid surface. To describe this phenomenon, simulations were conducted that modeled the movement of oil droplets along solid walls. The presence of surfactants at the oil droplet surface led to the most considerable migration velocity and displacement observed under Couette flow. The surfactants affect the interfacial tension, enhancing the momentum transfer from the water to the oil and reducing the oil−water slip. When the surfactant concentration is relatively low, the effects of Marangoni stress are overwhelmed by the flow field. Concerning Janus particles (JPs), they impede the migration of droplets. This is attributed to a weaker interaction between Janus particles and the solid wall in the presence of surfactants. Consequently, the coexistence of surfactants and JPs results in a more significant displacement than in bare oil because of the smaller interfacial tension.
Additionally, the anticipation of fingering is crucial when addressing the displacement of immiscible fluids, as viscous fingering is prevalent in the oil and gas industry. This phenomenon occurs when oil and water move together in porous media close to reservoir wells, but water moves more readily than oil, leading to instabilities and the development of water fingers. While the viscous fingering phenomenon has been investigated for bare oil-water systems, the effects of surfactants or particles at the oil-water interface are yet to receive detailed investigation. Our simulations involve injecting water into an oil-filled channel to displace the oil with different types of NPs (including homogeneous particles and amphiphilic JPs), surfactants, or both JPs and surfactants before pushing the oil forward. JPs exhibiting a dual chemistry, with a polar face featuring a contact angle θ_P and an apolar compartment characterized by a contact angle θ_A, inherently possess an amphiphilic nature. The degree of amphiphilicity of JPs, denoted by Δθ=(θ_A-θ_P)/2, is controlled during the computations so that the range is from zero to 47.8°. The computation results suggest that an oil droplet attached to a channel wall exhibits the greatest migration distance when only surfactants are presented at the interface. The droplet demonstrates the least migration in the presence of particles alone, regardless of their wetting characteristics. The re-arrangement of the surfactant molecules and the NPs at the interface, generating weak Marangoni stresses, plays a role in the motion of the drop. The instability of the oil-water meniscus arises from water flow, resulting in fingering phenomena before the eventual full detachment of the oil drop from the channel wall. The appearance of fingers depends on the presence of NPs or surfactants, with the particles leading to longer fingers at the same fluid velocity. A critical contact angle appears to dictate the detachment of the oil droplet from the channel wall, and this remains constant whether particles are present or not. When the oil drop detaches fully from the channel walls, the particles and the surfactant molecules reorient on the interface, modifying the shape of the droplet.
Last but not least, investigating the desorption energy of NPs at the interface of oil and water phases in the presence of surfactants and/or NPs is vital for numerous applications across various industries. The presence of surfactants alters the interfacial properties, potentially influencing the desorption behavior of NPs. Both surfactants and NP have the potential to influence the behavior of interfacial properties, which lead to variations in the desorption energy required for NPs to detach from the interface. In addition, while many studies overlook specific interactions between particles and surfactants, it is notable that surfactants can indeed interact with the surface of NPs, potentially influencing desorption energy. Hence, careful consideration of the interplay between surfactants and NPs is essential in the study of desorption energy. The thesis seeks to gain insights into the energy needed to detach a NP (including the homogeneous particle and amphiphilic JP) from the oil-water interface with the presence of NPs, and surfactants, as well as their coexistence, into one of the homogeneous phases. When examining the desorption energies of individual particles with the center of NPs as they traverse the oil-water interface, it is observed that the presence of surfactants can reduce the energy requirement, whereas the presence of NPs may lead to a challenging trend. Regarding the effect of NP type, hydrophilic NPs are favored in the water phase due to their lower desorption energy compared to the significantly higher energy required in the oil phase. Conversely, hydrophobic NPs exhibit an opposite trend. In the case of Janus NPs, the desorption energy level lies between those of homogeneous NPs.
These findings will be focused on exploring the mechanism of stabilizing the interface between oil and water using NPs and surfactants, as well as the potential implications for tuning oil and water interfacial behavior
A Punishment Beyond Sentencing: Incarceration's Impact on Overall Quality of Health
No full textIn this dissertation I use Agnew’s General Strain Theory as a frame and examine whether criminal justice involvement, specifically ever being arrested, ever incarcerated, or paroled, is related to the overall physical health and livelihood. I test for relationships between criminal justice involvement and two outcomes: respondents’ self-reported fair/poor health and respondent stating that health affects their ability to work. A third outcome tests how criminal justice involvement in the last 12 months, specifically being paroled and being arrested, is related to using negative coping strategies, specifically illicit substance use and prescription opioid misuse, over this time period.
First, I examine how the self-reported health is shaped by incarceration using the National Longitudinal Survey of Youth (NLSY97). I find that ever being incarcerated has a significant negative effect on self-reported health outcomes; however, once access to health insurance and other social inclusion variables are introduced, the probability of reporting worse health outcomes is reversed, suggesting that there is an underlying relationship between having worse health and the social inclusion variables, and that the relationship between ever being incarcerated and worse health is mitigated by social inclusion factors. Being socially excluded is associated with worse self-reported health. These findings provide support for the health disenfranchisement hypothesis as well as the social exclusion hypothesis.
Next, I use the NLSY97 to examine the independent effects that incarceration and social inclusion have on respondents’ self-reported ability to work. As predicted, I find that the experience of being arrested is associated with a higher likelihood of respondents reporting that their health affected their ability to work. While the number of incarcerations and arrests in the last 12 months variables both share positive bivariate relationships with reporting that their health affects their ability to work, these relationships do not hold once demographic and social inclusion variables are included in the models. I also find that African Americans and females have a higher likelihood to self-report an inability to work due to their health. These findings for African Americans and females support the social exclusion hypothesis, but the relationship only holds for women once the social inclusion variables are included.
Last, I use the National Survey of Drug Usage and Health (NSDUH, 2021) to test whether being arrested and being on parole is associated with using the negative coping strategies of illicit substance use and prescription opioid misuse, while testing for the moderating effects of social inclusion factors. I find that experiencing an arrest and being paroled in the last 12 months are both associated with using negative coping strategies, which lends support to the social exclusion hypothesis. I find support for Agnew’s general strain theory in that formerly incarcerated people are more likely to be vulnerable to negative stimuli, resulting in higher odds of using negative coping strategies. This aspect of general strain theory supports social exclusion due to the limited life chances that formerly incarcerated people can cultivate during the reentry process.
Overall, this dissertation provides evidence that criminal justice involvement negatively affects the life chances and coping behaviors of formerly incarcerated or paroled people. However, social inclusion factors provide a salient countering influence to the negative health effects of being arrested, incarcerated, or paroled.
Keywords: Incarceration, arrest, parole, health, life chances, stigma, reentry, health care access, negative coping strategies, opioid, cocaine, methamphetamine, heroin, social inclusion, social exclusio
Examining The Impact Of Different Types Of Social Media Influencers On Attitudes, Trust & Purchase Intent: Travel User Generated Content
Get PDFABSTRACT
This study examines the effects of Instagram nano, micro, and macro influencers on consumer’s brand and influencer attitudes, trust and purchase intent toward travel-related content. Using the Persuasion Knowledge Model as a theoretical framework, this study analyzes how the presence or absence of sponsorship disclosures within Instagram posts might mediate and affect consumer’s attitudes and behaviors. The purpose of the study is to add to influencer marketing literature by exploring how consumer attitudes and behaviors may vary across types of influencers on the presence or absence of sponsorship disclosure. The study employs a 3 (influencer type: macro-influencer vs. micro-influencer vs. nano-influencers) x 2 (sponsorship disclosure: present vs. not present) between-subjects design experiment. Through a experiment, University of Oklahoma students were randomly assigned to 1 out of 6 possible groups. Participants were shown a fictitious Instagram influencer post promoting a beach resort. After exposure to the Instagram post, trust, attitudes, and purchase intent effects were analyzed. The presence of statistically non-significant results may indicate an intricate relationship between influencers and individuals. This potential complex relationship might benefit from further research incorporation the role of parasocial relationships. Understanding real life relationships between influencers and followers might result in more accurate results. Despite this, the study found that there is an interaction effect between influencer type and sponsorship disclosure on participant attitudes toward influencers. A two-way multivariate analysis of variance (MANOVA) showed that non-sponsored posts from micro-influencers received significantly more positive attitudes toward the influencers than sponsored posts from the same influencer type. Additionally, attitudes toward nano-influencers in sponsored and non-sponsored posts were similar. Also, sponsored posts coming from macro influencers showed more positive attitudes toward the influencer in comparison to non-sponsored posts. Furthermore, the study found that individuals who were exposed to sponsored Instagram posts had low levels of purchase intent. An independent sample T-test was conducted to compare the mean levels of purchase intent between participants exposed to sponsored and non-sponsored Instagram posts. Results showed that participants exhibited both low purchase intent levels toward sponsored posts and lower purchase intent levels in comparison to non-sponsored posts. Finally, an independent sample T-test was conducted to compare the mean levels of influencer attitudes toward sponsored and non-sponsored posts. Results did not only show that participants exhibited negative attitudes toward influencers in sponsored posts, but also that influencer attitude levels were more positive in non-sponsored post