Questions related to Intravenous Therapy
IV and DV have significant correlation (.345**), and so is the case with IV and MV (-.388**)and MV and DV (-.264**). Mediation model is run on IBM SPSS and shows coefficient for direct and total effect as 0.084 and 0.101 respectively, and coefficient for indirect effect is 0.017 with BootLLCI (-0.008) and BootULCI (0.054) having values with different signs.
Please suggest ?
Dehydroascorbic acid crosses the blood brain barrier and preliminary research shows it has a neuroprotective function.
I am writing my thesis and I am a bit confused in reporting my covariate. my
IV are: brain stimulation, and time
DV: rate of percieved exertion (RPE)
I run repeated measures Anova, but want to control for sex so I put it as covariate, and I found significant value for sex which makes it covariate. However, no significant effect of IV on DV. shall I still report the sex effect? or just say: "no significant effect was found of IV on DV while controlling for sex".
I will appreciate your advices and answers.
Thank you very much
#Statistics #neuroscience #Covariance
I am conducting some research, and I have 1 IV, 1 DV and four covariates; I was initially planning on carrying out an ANCOVA, but it turns out that the DV will have to be binary( yes, no ) answers. Any suggestions on alternative analysis?
Hello, as I was doing my data analysis, I came across what I read as a suppression effect. One of my IV which is low but positively correlated with my DV seems to have a negative effect during multiple regression analysis. But how can I explain it? What does it say about the relationship? Is the correlation false? What should I do?
Hello, I am currently conducting a moderated mediation analysis in AMOS and want to mean centre my IV and moderator. To calculate the mean, do I use the original dataset or the dataset where I have removed some items with low factor loadings.
Hello to all,
I do not know how to make mediation and moderation with covariates in SPSS with a categorical IV and DV (both have two categories 0 and 1). I am doing a between-subject design.
Thanks in advance for our help;
I am trying to understand if the administration route can influence this AAV serotype's expression in the brain.
We are going to induce intracerebral hemorrhage by using collagenase type VII (Sigma). We used to do this before using collagenase type IV by dissolving it in PBS. However, Sigma emphasized that we must dissolve collagenase type VII in TESCA buffer. Is it possible to dissolve this collagenase in PBS as well?
Thank you and by best regards,
The BET analysis of my synthesized ZIF-8 at 77K gives a type IV like isotherm. The sample was degassed at 150 degree C for 8 hours. The BET surface area of ZIF-8 I got is around 1800 m2/g, which is almost similar with the previous literature reported. The t-plot data also gives micropore volume of 0.7 cc/g. But the isotherm shows a noticeable hysteresis loop from relative pressure range around 0.4 to 0.8. Is it possible for ZIF-8 to show a type IV like isotherm as in most of the literature ZIF-8 MOF is reported as microporous MOF. Is there something wrong or there may be some reason. Please suggest
Our IV has 4 quadrants which all are not significantly correlated with the DV but the hypothesis include moderation, backed by literature.
Can we still proceed to moderation analysis?
I've been looking into the effects of pre-order promotion design on pre-order sales of videogames.
My IV"s are binary, 1 moderator (AAApublisher, referring to firm size) is binary, and my second moderator (Sentiment, low is negative, high is positive) is continuous. Sentiment has some missing values. My DV is number of pre-order sales, which I've had to log-transform due to its distribution.
I've attached my conceptual model.
When I make a simple linear regression model with only my IV's, the results suggest that they've got decent explanatory power of the DV, but when I add my moderators, these main effects disappear. As I understand, these moderators (and not my IV's) are actually the cause of the changes in my DV.
Looking forward to hearing your thoughts on these results. Thanks!
Family member with diabetic retinopathy and who has poor vision from blood being trapped in the eyes stayed at a hospital on two separate occasions and only received a sodium chloride saline IV bag and was able to see much better because the blood had mostly receded from the eyes. He also had a potassium sodium imbalance, high potassium of 5.4 and low sodium of 121. Is there a way to replicate this treatment at home to test and see or something similar to offer relief for this condition? He stayed at the hospital the third time but did not receive the IV solution and did not have an improvement what would be the reason for the blood to almost recede completely? Looking forward to your responses. Thank you and have a blessed day
how do we categorize SES based on the Kuppuswamy scale then make it into three categories like Low, Medium, and High?
Since scores for kuppuswamy sale are.
Socioeconomic class Total score I Upper 26‑29 II Upper middle 16‑25 III Lower middle 11‑15 IV Upper lower 5‑10 V Lower below 5
this can be recoded into three? any reference?
I got set of data that includes:
Gender: categorical (classified as IV in jasp)
Ethnicity: categorical (classified as IV in jasp)
Congruent: continuous data (classified as DV in jasp)
Incongruent: continuous data (classified as DV in jasp)
I have been asked the following questions:
Is there a significant interaction between ethnicity and implicit association?
I am struggling to choose the correct test; I am trying ANOVA but actually I don’t know what I should measure to answer the question!
Is it the interaction between Ethnicity and gender? What about congruency data?
i have multiple dv's and m ultiple IV's ,research paper and articles which i have read have suggested me to use multivariate multiple regression. but the syntax in spss of multivariate multiple regression and MACOVA seems to be same. Analyze>Genral linear>multivariate.
i am putting Dependent variable under the dependent variables category and independent variables under the Covariates category as variables needed to be in nominal scale ,if they have to be put under the Fixed Factor category.
please help and suggest my all the DV's and All the IV's are measured on 5 point likert scale
Currently, I am working on project of "CHROMIUM (VI) REMOVAL FROM WASTEWATER USING ACTIVATED CARBON DERIVED FROM ALGAE". If anyone working on the same project, can you please share details of what you had done in this regard. Besides this, if you had used/tried other activated carbon including but not limited to tea waste, tangerine peel, bovine gut, sunflower seeds, can you please share details regarding the drawbacks of those activated carbon that you had used in your project?
My framework IV on DV through Mediation. Is any requirement that must measure direct effect of IV on DV? Because I wanted to prove that Mediation makes a positive significan and never claim that direct effect is significant.
During a process of polyester staple fiber production, while the IV (intrinsic viscosity) of PET granules is within 0.63-0.64 dl/g, the IV of as-spun fiber output from spinneret is around 0.66-0.67 dl/g.
What are the reasons for such growth in IV?
Does IV rise have any negative effect on polyester fiber quality?
For some time now, I cannot figure out what is going on with a moderation model (model 1) with a covariate that I'm running.
IV = categorical, two groups, dummy coded (lockdown (T2) or no lockdown condition (T1))
DV = continuous (mental illness)
M = continuous (resilience)
Covariate = gender (dummy coded)
I set the conditioning values at 16th, 50th and 84th percentiles. Further, I mean centred the continuous variables.
The overall model is significant.
When I run the data for visualizing the conditional effect of the focal predictor, I get a graph that I do not think is right. Resilience is pictured on the x-axis, mental illness on the y-axis, and the lockdown or no lockdown condition is noted as two variables (0 and 1) on the upper right side of the graph. I always thought that that is where the moderator should be, not the IV. That's why I'm confused.
Can somebody explain to me what is going on or what I perhaps should do differently in the analysis? A big thank you in advance :)
Hello. there's something that has been confusing me.
I wanted to conduct a tensile test for Polycarbonate materials printed with FDM 3D Printer. In the ASTM D638 standard, there's five types of specimen, I to V. the standard recommends the use of type I specimen whenever possible. However, when I look up what other researcher had use in their respective research. Some, if not, the majority of papers that I read uses the type IV specimen. Despite it being stated that the type IV specimen is used only when we want to compare different rigidity, or to be used for non-rigid materials. Most of the papers that I read tested rigid materials such as PLA, ABS, or even PC itself. Is there any reason for that? And is it still valid? Because I also would like to use the Type IV if its allowed to save cost and time.
Please let me know If you know any things about my question. Recently I have conducted a mediation analysis using PROCESS model 4 macro software and with the bootstrapping method (Hayes, 2017). As you know, in these conditions, we have three paths, A, B, C
path A is between IV and M(mediator), path B is between Mand DV, and path C is between IV and DV when M also is in the model. Please let me know when we can say the indirect effect is significant? should all three paths be significant before? for example when path C isn't significant, and or two paths C and A aren't significant, we can say the indirect effect is significant (I know when zero isn't in the confidence interval (BootLLCI and BootULCI), we have the significant indirect effect). what about when we have two and or three mediators in the model?
All the best,
I applied Ordinal Logistic Regression as my main statistical model, because my response variable is 7 Point-Likert Scale data.
After testing for Goodness of Fit using AIC, i got my best fit model, including 4 independent variables (3 explanatory and 1 factor variable).
However, I encounter 1 negative coefficient value (0.44 odds) of 1 explanatory variable (all explanatory variables are also 7 point Likert-scale).
My theoretical assumption is simple: the more frequency of explanatory variables (engage in activities) happen, the higher impact score on response variable (mutual understanding)
That's why I am confused when 1 independent variable has negative coefficient.
In this case, how should I interpret this IV?
Thank you very much,
I have a query regarding which type of regression analysis to use for my study. I have used a scale (dependent variable) that contains 9-items and each item is marked on a 5-point Likert scale. Scores of each item are summed and ranges from 9 to 45. Higher score indicates the respondent has more characteristics of that construct.
Similarly, there are two independent variable. One IV has 20-items and each item is marked on a 4-point Likert scale. Score ranges from 20 to 80. Second IV has 7-items and each item is marked on a 5-point Likert scale. Score ranges from 7 to 28.
The reviewer has suggested me to use non-parametric tests since my data is ordinal. However, previous studies have used Multiple Linear Regression using similar types of constructs.
Which type of regression analysis is appropriate in this case - ordinal regression or multiple linear regression? Any literature explaining this would be highly useful.
Please mention name of the regression analysis suggested.
I am currently conducting a study for my master's thesis about the effect of empathic concern and personal distress on purchase intentions (two of the subscales of the IRI by Davis as IV's) and WTP for sustainable apparel, with attitudes towards sustainable apparel as a mediator. I have two questions regarding the analysis. Besides the two IV's, the mediator and the two DV's I also include 6 different control variables.
1. Does anyone know if the subscales of the IRI are reflective or formative?
2. A researcher argued that as the IVs are formative (not sure if that's the case though?) and the model is complex, PLS needs to be used. Would you agree with that? Or do you think OLS is sufficient? Would it make a difference if the IVs were reflective and not formative?
Kind regards and thanks a lot in advance.
What synthesis approach would be more appropriate to obtain palladium oxide nanoparticles using potassium hexabromopalladate (IV) (K2PdBr6 ) as precursor material?
There have 3 independent variables which are knowledge, perception, and awareness. My dependent variable is buying decisions. The question in my questionnaire only has demographic and IV questions. But the IV questions represent their buying decision. So I still need to create dependent variable question? My hypothesis is looking for the different status (worker and student) of their buying decision.
I ran a logistic regression with continious IV in SPSS. In the table "variables in the equation" one variable is missing (despite using entry method) and without any message from SPSS . When browsing through the web I understood that this might happen due to collinearity. However Collinearity diagnostics did not return a clear sign for it. Highest VIF-values are 6.1 and the highest Conditionindex is 21.1.
So my question are:
1. Is my regression model still valid despite SPSS dropping one variable?
2. Are there other reasons than collinearity why the IV is missing in the model.
I have 3 DVs and 5 predictor variables (technically, only 1 IV and the other 4 are controls).
I ran it on Stata and all seems to have worked (since mathematically, all predictor variables are treated the same way), but technically I only identify one of my predictor variables as the IV of interest. I would rather stay away from SEM. Thank you so much!
Say we have a Multidimensional variable (having 4 dimensions) that acts as an Independent variable (IV) and want to measure its impact on a uni-dimensional variable (dependent variable).
Would it be okay if we take a composite score of IV (= sum of scores on all the 4 dimensions of it) and run the analysis while the DV is a single dimensional construct?
In other words, one variable (IV) has second order factor structure while another variable (DV) just has first order factor structure...
Thank you in advance!
I am working on Perovskite solar cell. We have a KEITHLEY 2450 SMU at our lab. I want to measure IV characteristics of solar device with FTO as front electrode and Ag as back electrode material using this SMU. My question is; which type of clip (for example aligator clip etc.) needed to make good contacts with electrodes in my thin film solar device.
Scenario : You are shifting your patient & the epidural catheter got broken at the filter end. You troubleshoot but to no avail it is working. The surgery and patient characteristics are : ASA 1, obese male & he underwent open radical nephrectomy with a loin incision with only one functioning kidney.
What would be your alternatives to manage pain in a resource limited setting? Will IV tramadol, with IV PCM (mulitmodal) produce similar pain scores in a patient receiviing epidural morphine alone?
Does Beta Value found out using AMOS Software shows the influence of one Independent Variable (IV) on Dependent Variable (DV)
If I have the dose of some drugs taken intraperitoneal , how can I convert it to an intravenous dose both are given to rats ? would it be given with the same dosing frequency ? the only available data in literature is the oral bio-availability (F=49%)
I'm currently designing the study for my dissertation. Originally I was just going to design a plain phenomenological study where I select participants and conduct a structured interview with all of them and find the pattern and then discuss the final result and say how does IV impact DV based on the pattern of the qualitative data.
After talking to many professors from other psychology departments, they gave me cool suggestions saying it would be more convincing if I could conduct a comparative study where one group of participants are selected with the IV and another group of participants selected without the IV and ask these two groups the same set of interview questions and cross-compare the patterns of these two groups' interview results.
My questions are :1. Is there a specific method name for this research design? 2. How should I frame it in writing the methodology chapter if it doesn't have an already existed term?
First of all, I really appreciate that you take time to help me out.
Info about my conceptual model:
Moderator = Social Capital
IV = entrepreneurial intention
DV = entrepreneurial behavior
Hypothesis 1: There is a positive relationship between entrepreneurial intentions and entrepreneurial behaviour.
Hypothesis 2: An increase in social capital positively affects entrepreneurial intentions.
Hypothesis 3: Social capital positively moderates the relationship between entrepreneurial intentions and entrepreneurial behaviour
My question is how to test H2 because I find many tutorials on how I test H1 and H3 with Structural equation modeling (SEM) but I can't find anything about H2. Could you help me out?
I'm currently writing on a dissertation, c.40 pages, and I'm left with no academic support due to the holidays now.
My topic in international relations is very specific and something about which there exists only scant literature. In fact, most of that literature redefines the phenomenon (very contentious that it does exist), but I want to explain how it actually emerges, the reasons why it happened. The work will be done comparatively with two case studies and process tracing. The phenomenon logically is the dependent variable.
From studying the two cases I have gained a good understanding of the drivers behind the phenomenon, and those drivers explain the emergence of the phenomenon in both cases. So the drivers are the independent variables. What I know so far helps me to bridge the gap between IV and DV; that is, explaining the mechanism that led to the occurrence of the DV.
However, what stuck me know from reading guides on how to write dissertations (or Bachelor/Master thesis) is that they all presuppose a 'theoretical framework', or a 'theory' that drives the research question. Apart from conceptualising how to think and measure the DV, I don't really know what else I should add. There is no literature/theory that would explain the DV. Why can't I just proceed as I planned?
Would appreciate any help!
I am now conducting a quantitative study utilizing a descriptive correlational methodology. I am also employing an Independent Variable and Dependent Variable with a Moderating Variable as the third variable. Can you assist me to determine what statistical treatment is essential to investigate the correlation of the IV towards the DV with the intervention of the MV?
Thank you very much!
I have a question:
What statistical analysis should I do when I have:
- IV: Sustainable Campagains
- DV: Customer choice in purchasing sustainable t-shirts
- Mediator : Social Pressure
- Moderator: Quality perception
Can I do ANOVA? Or I have to do regression as well?
I am trying to find a way to take into account local buckling in Class IV sections in Abaqus. This can be done in SAFIR via recalculating the material plasticity under compression, like Franssen et al. in "Effective stress method to be used in beam finite elements to take local instabilities into account" but I cannot find a way to implement a similar method in Abaqus.
I would appreciate very much if someone can give me a hint about this matter or point me in the right direction with some publication.
Thanks in advanced,
I am writing a final year project where my independent variable is employee well-being and my dependent variable is happiness. the factors of my IV are social well-being, physical well-being, and psychological well-being.
Are there other factors that I could add?
Also, I would love to gain more advice and insights.
Please consider 3 variables:
- DV - dependent
- IV - independent
- MV - mediator
Am using PLS-SEM for analysis. And this is my problem
IV --> DV is a negative, insignificant relationship (acceptable according to theory)
But on introducing a mediator:
IV --> MV --> DV, I get the following:
1) IV --> MV: significant negative (expected according to theory)
2) MV --> DV: significant positive (expected according to theory)
3) IV --> DV: significant positive (certainly not expected)
Variance inflation factors have been checked and are not an issue.
I have been reading about suppression, but I am not sure that is the case here.
Would be most grateful for any suggestions.
I am working in perovskite photovoltaic lab. We are fabricating carbon based monolithic perovskite solar cell. The structure of our cell is FTO/c-TiO2/m-TiO2/ZrO2/C and perovskite infiltration. We have fabricated the solar but we are not getting ideal IV curve which also results in low efficiency. How to improve the fill factor and get better IV curve?
My moderator is Social Capital. A questionnaire (5 points Likert scale) of 16 questions is used to measure it. (IV = entrepreneurial intentions, DV= entrepreneurial behaviour, M= Social Capital)
My research focuses on if respondents acquired social capital through a (former) membership at a student association. So after the 16 SC questions, the last question was: Have you gained social capital through (former) membership of a student association? (also a 5 point Likert scale)
Therefore, my question is, how do I compute/combine these 2 different aspects into one value for each respondent so I can use it to test my hypothesis.
Thank you in advance!!
In a mediation analysis, the direct effect between DV and IV is non-significant. However, the indirect effect becomes significant when the mediating variable comes into play. I often read in sources that for mediation analysis there must be a relationship between DV and IV. According to my logic, if a variable makes a normally non-significant relationship significant, it should be possible to talk about a significant mediating effect there. How would you interpret this? And would such a situation pose a problem? Thank you in advance for your answers along with your source suggestion on the subject.
I am using a questionnaire of Likert scale as a Quantitative method based on analyzing data intervals collected from a survey of a scale (strongly agree, agree, neutral, disagree, strongly disagree).
I have classified my independent variable as follow, but finding difficulty in terms of identifying my dependent variable
IV: Using smart system to measure its effectiveness if it was to be implemented
Now my question is, what should be a proper dependent variable? Can it be individual's opinions gathered from the survey?
Looking to start utilising ADIVA scores as part of a unit wide approach go identify patients for whom gaining IV access is "problematic" in an attempt preferentially move high DIVA score patients towards initial attempts via ultrasound guided cannulation. Has anyone had significant experience with this ? are the DIVA scoring methods effective or have you had to develop your own ?
I am working with two sets of colleagues on research in which we have conducted multiple regression analyses. In one project, there are two independent variables (IVs), one of which is categorical (it's gender) and the other IV is continuous (age). In the other project, there are 11 IVs (they are all significant after backward elimination of several other IVs), most of which are categorical.
I am not sure whether we should be reporting Cohen's f-square as an indication of effect size or something else when describing our results. My main concern is whether the presence of categorical variables would render f-square inappropriate.
Can anyone help, please?
Hey there! I'm currently writing my BSc thesis, and I really need some help with SPSS and its interpretation.
I'm doing research into the effects of gamification (dichotomy) on learning, with motivation as a mediator, and gender; interest, and experience with gamification as moderators.
I was advised to apply PROCESS Model 5 to the data (a moderated mediation model), and run separate analyses for each moderator whilst including the others as covariates. Now, I thus have three different analyses, with three different models.
With regards to mediation - the effect of gamification on motivation remains the same in all three models, regardless of which moderator is applied. However, I cannot report the effect of motivation on learning, since in each model it is different - e.g. b = 0.074, t(57) = 2.29, p=0.202 OR e.g. b = 0.078, t(57) = 2.44, p = 0.018.
With regards to moderation, I assume I must only report the moderator and not the covariates..?
One last thing... How does it affect the data when my IV is dichotomous, and can I avoid this? The IV dichotomy was first coded as 0 - gamification and 1 - traditional education, and now I swapped them around. If I compare the data, some coefficients only gained a minus (-) sign in front of it, whereas e.g. 'constant' and moderators change numbers.
I would be so grateful if you could help me out, even if it's only an answer to one question!
b = -0.42
Sy = 0.95
S x = 0.99
I need Beta. But the predictor (x) is already standardized. May I nevertheless use this formula?
beta = b * ( S x / Sy)
Thanks for help
I have a 1970s publication citing a section in this 1950 version of Ham on the saphenous vein. I have tried to obtain a photocopy of the relevant section via the publisher and various libraries (including my own at UCL and the British Library) but without success. I need to confirm that reference to a particular part of the text is true. Any help is much appreciated.
I am looking at whether stress levels reduce from time point 1 to time point 2 when engaging in recommended resources.
DV's - stress levels from time point 1 and time point 2
IV's - engaged in resource 1 and resource 2
I have 2 IV:
- Scale data of skills achievement by our University alumni (They rate from 1 to 7 their acquired skills at University)
- Gender (Already converted it to a dummy variable)
And I have 1 DV:
- Job finding difficulty that has 5 options: (Very Easy, Easy, Neither Easy nor Difficult, Difficult, Very Difficult).
I have 2 questions that I need your assistance with:
1. Can I use a dummy variable with 0, 1, 2, 3, 4, 5 values for my DV?
2. My first IV is not normally distributed, can I still run this regression with it?
I am trying to see the contribution of gender or any specific skills on job finding difficulty.
Please answer as clear and as short as possible.
Hi to all,
I am using simultaneous equations for my study and I have to solve the possible endogeneity problem.
In our study, we are trying to analyze factors that influence farmers' decisions to purchase seeds from different sources. We suspect the existence of endogeneity between seed source and seed variety purchase. And we don't have suitable instruments to solve this problem.
Hence, I am looking for an econometric tool to solve this problem when the dependent variable is binary (seed source) and continuous endogenous variable (varietal age) and suitable IV's are not availability.
STATA code, ivreg2h is not suitable to my condition seems as my dependent variable is binary
Hoping someone could help me to solve this issue.
I have incorporated a single impurity in 1*5*1 supercell of GaN. For 4 different atoms from group IV I have found the same lattice constant. Is it normal?
I am very new to SPSS and I am trying to understand how and what type of statistical test that I should use for moderation analysis. For the first part of analysis, I will be doing Pearson correlation analysis to assess the relationship between the IV and DV. However, Im not too sure about the next step to analyse the effect of the moderator variable in the relationship between IV and DV. Please find the attached conceptual framework of my thesis.Can someone guide me on the exact next step to be taken ?
I determined the surface area of CeO2 modified with MgO as 1 m2/g. The isotherms showed significant hysteresis loop (type IV), indicating mesoporous structure, which is strange to me. Can anyone please explain why I am getting hysteresis at such small BET surface area of 1 m2/g?
Also, the pore size is significantly large, about 15 nm.
My research model is based on the SOR model on the attachment. In my experiment, I display two conditions: red and blue environmental stimuli, to each respondent. Then, I ask each respondent to rate their emotions and responses to both conditions. I know how to use SPSS PROCESS model 4 to test the mediating effect if IV is continuous, but I don't know to test the mediating effect of emotions when the IV is the dichotomous variable, namely, red and blue environment.
Thank you in advance!
I want to set each of a survey's phrase (question) in a way that reflects independent variable and dependent variable together .
for example : I have CSRD as a IV , and Competitive advantage as a DV , so i want to write phrases (questions) that combined the two variables at once not separately (i.e not phrases(questions) related to CSRD alone and phrases related to Competitive advantage alone).
so is this method correct ?,if it is true, what kind of statistic test may i conduct?
thank you .
I am setting up mitochondrial complex specific assays performed with a spectrophotometer. I am interested in setting up the assay to look at complex IV activity using the electron donor TMPD. Upon metabolism of TMPD by mitochondrial complex IV the amount of oxidized product of this chemical should increase and therefore the absorbance of the absorbance maxima of the oxidized TMPD. At what wavelength does the absorbance of reduced and oxidized TMPD differ maximally so that I can correlate amount of oxidized TMPD with complex IV activity?
A module contains large number of cells connected in different strings. We can extract the IV curve of entire module using flash IV simulator but I am interested particularly finding or extracting IV curve/characteristic of a single cell.
Is there any way of doing this?
Hi I want to conduct a SUR analysis on two regression models with the same IV but different DV to compare the contribution of a same IV to the two DVs, but the two DVs are different in terms of order of magnitudes. I wonder how I can standardize coefficients when conducting SUR analysis using the xtsur conmand in Stata, shall I just zscore everything before doing SUR analysis?
I am currently writing my master thesis about customer acceptance of service robots. I use SPSS Process model 9 (moderated mediation).
My IV is robot's voice type (human vs. robotic), mediator is robot's perceived credibility and DV is customer's service encounter evaluation. The a-path from IV to M is moderated by robot's appearance (humanlike vs. machinelike) and customer's extraversion.
In my model, I take different control variables/ covariates into account which are interesting in robotic research. I look at participant's age and gender and if they have had previous interaction with a robot to check if those variables have an influence on my DV. Age and gender seem to have no influence, but robot interaction does.
However, I also collected data on the participant's nationality/country of origin. Taking this variable into the model I have some interesting results, such as that German participants rated the service a lot lower than other nationalities.
Now my problem is, that my Independent variable voice type becomes slightly insignificant when I add nationality/country as a covariate (of course dummy coded). Before adding country the voice style’s p-value was 0,0598 and with the nationality/country, it is at p=0,1007 ( I am using a p-value of <0.1).
My second concern is, that my nationalities are not evenly distributed. I have 501 participants in total and 259 germans (51,7%) and the second-highest is the USA (49 people, 9,8%), Taiwan (8%), the Netherlands (6,4%), UK (4%), Franke (3,2%) and other countries (17%).
So I am wondering if my results can be representative if more than half of the respondent’s nationalities are germans? Therefore, I was wondering if it would be recommend to take nationality into account as a control variable or if I should leave it out, because then also the c' path from voice type to service encounter evaluation would remain significant at p= 0,0598.
Thank you for your help! Best, Nina