I'm writing my master's thesis on the investigating of delay factors on building construction projects. My independent variables (8) is the delay factors (materials-related, manpower-related, equipment-related, project management-related, consultant-related, owner-related, governmental-related and external-related). The dependent variables is delay in construction for frequency in occurence and degree of severity. which statistical test will be appropriate to apply in SPSS?
Null Hypothesis (H0): There isn’t a significant statistical differences in frequency and severity between causes of delays caused by delay group.
Alternative Hypothesis (H1): There is significant statistical differences in frequency and severity between causes of delays caused by delay group.
Is One-Way ANOVA is applicable for degree of severity and frequency of occurence?
Salam Dalal Alwedaei
If you are aiming to identify the differences between groups' mean, Yes, One-Way ANOVA is applicable for both the degree of severity and the frequency of occurrence of delays caused by delay groups.
you can also determine if there are significant statistical differences in the degree of severity and frequency of occurrence of delays caused by different delay groups. you need to check the assumptions of One-Way ANOVA, such as the normality of the data and homogeneity of variances, before conducting the analysis to ensure the validity of the results. all the best
Dalai Alwedaei:
One Way ANOVA is applied for mean of group.
For this test if necessary normality and homokedasticity. Un RStudio is very easy applied It.
If H0 rejected, you compare one to one mean for determinate which is differents, the next step is eliminate It.
Good look in your Thesis.
Axel.
Hello Dalal Alwedaei. Regarding the dependent variable, you said:
The dependent variables is delay in construction for frequency in occurence and degree of severity.
I may be misunderstanding, but that sounds like two separate variables, delay (presumably some measure of time) and degree of severity. Please clarify.
And still regarding the DV(s), is it conventional, and sensible, and defensible to use means and SDs when reporting descriptive statistics? If not, why not?
Finally, tell us more about the 8 explanatory variables. Specifically, are they quantitative or categorical? Do you intend to include the first-order terms only, or do you want to include product terms (interactions) or polynomial terms, etc.? And what is your sample size? Thanks for clarifying all of this.
Before getting into possible statistical procedures (and Bruce Weaver 's point about what the variables are and if there are two DVs), I have some questions about the null hypothesis. "Null Hypothesis (H0): There isn’t a significant statistical differences in frequency and severity between causes of delays caused by delay group." First, the null is usually NOT whether there is a significant difference, but that there is absolutely no difference in the population. Can you verify this is what you mean? Otherwise define what you mean by significant. Further, please clarify what you mean "between causes of delays caused by ..." Is your null that in the population there are no differences in both the frequency and the severity variables? And frequency is presumably a count variable, but you also talk about delay which is presumably a time variable. And finally, do you mean each of these eight has zero influence, the eight as group, etc. As Bruce says, other aspects (e.g., are these 8 predictor variables and how so they relate) could also be made clearer if you wish to useful feedback.
As Bruce mentioned the type of the variables might be interesting.
Another idea: Maybe a multivariate analysis, like PCA, could be interesting to identify how different variables impact the delay and wether some of the groups have similar effects.
You can use mixed effect linear model for analyzing said dependent and independent variables.
- Badges
- Science topic
- Similar topics
- Biological Science
- Zoology
- Ornithology
- birds
- Scopus