There are fundamental questions to aske here:

1. Did the research use validated questions/instruments (if yes and reliability checks, factor or regression followed by structural equation modelling was undertaken) you can therefore gauge the validity by determination of goodness of fit and comparisons with the earlier validated model

2. If the questions were not previously validated then some traditional processes for validation involve approaches such as 

(a) take results and then either apply the model/simulate the model to observe outcomes (or do not apply but simulate/predict and compare outcomes with simulation/prediction), then either follow the process (or simultaneously) by determining model fit

(b) undertake numerous determinations over time for the model under different circumstances, repeating (a) above to examine any circumstances where the model does not have good fit (the development of many psychometric measures followed this approach), this will allow you to determine whether ther are constraints on validity

Thanks Mr.Robert i have gone through the first process that you have mentioned, data collected through a survey instrument  from 408 respondents , structural equation modeling technique applied  and all the measurements indices for goodness of fit are accordingly to the threshold values. Now do you think that we need any expert opinion for further validation, or the process done is enough?

After using structural equation model, you should present your model to some expert and ask they opinion about the feasibility of the model (expert consensus).one of the good statistics tool that can help you is chi square.

@ Muhammad

 You did not respond to the very pertinent query from Robert "Did the research use validated questions/instrument", on which answer the further actions follow. I do expect you have set up a questionnaire on your own (since you are talking of a conceptual model).  If yes, then the answer to Robert's query is No.  If No, then the question of validation of the questionnaire should have been carried out by erstwhile user(s).

Alireza contemplated the rest of the action. RG would offer you several of them. it would be a pleasure to help you since I too formulated and validated a conceptual model.  in the alternative you could still use the methods advised by Ronert ie,  "

Apply the model/simulate the model to observe outcomes and compare outcomes with simulation/prediction), then  (or simultaneously) by determining model fit

Firstly, Use Factor analysis to establish the  construct Validity of the concepts

Secondly, Use Recursive Causal Models to test if the theoretical model is consistent with reality

Thirdly,, Use Cross Validation (Over boundary or over time)

Thanks Abdulrazzak,and Mr. Priyavrat  the instrument used in this study is such that out of 61 items, the first 18 items adopted from another researcher , while rest of the items are self developed. as concern the factor analysis i did principal component analysis and the value of KMO is 0.712, and value Barlett spherisity is also significant. content validity cheked through expert opinion from expert s of the filed.  while for model fitness i checked , value of CFI, RAMSEA, and Ch-sq/df .  Values of these indices are with in the threshold values. while i have not checked any simulation or outcome.

You have already done the surveys and administered them to  a sample of  school leaders. After your initial analysis, you need to perform a factor analysis to identify the constructs or dimensions.  In addition, make sure you do a reliability analysis to determine the reliability coefficient for the entire instrument as well as each construct. A reliability coefficient of .80 or above is considered good. You may discard items with very low correlation coefficients.

 Mr. Henry thanks that you added a lot, i have checked reliability coefficient for all three variables and its sub-dimensions and values are above .80, while i have checked AVE, and CR values and these values are satisfactory, while factor loading is above .50 for each dimension Actually what i mean for validation is the perfectness of this conceptual framework, that how i will ensure that this conceptual framework is useful to contribute in field.

A Statistical process determines Validity. Validation is necessitated when there is the absence of a mathematical or experimental procedure, and a kind of procedure is developed based on skills to get reproducible results. Thus, you appear to have ensured validity already. Now, your concern for validation is genuine to assure the perfectness of this conceptual framework.

Good going.

Further, as Abdulrazzak recommended-  Cross Validation (Over boundary or over time), you proposed - expert opinion and/or Robert advocated: applying the model/simulate the model to  compare outcomes with simulation/prediction)

Further as Robert advised - undertake numerous determinations over time and under different circumstances, repeating  to examine any circumstances where the model does not have good fit, to determine whether ther are constraints on validity?

Though only one of these is essential, but in Re-search... Carry on  (Bollywood movie Lage Raho Munnabhai MBBS) ;-)

Mr.Priyavrat, i like your last comments, but i am not as great as sanjay dutt, any how will continue further discussion on various topics.

Yes you need an expert in the area of statistics to walk with you

Absolutely right Lillian,

"To call in the statistician after the experiment is done may be no more than asking him to perform a post-mortem examination: he may be able to say what the experiment died of." Ronald Fisher (1938)

Though Benjamin Disraeli said: 'There are three types of lies -- lies, damn lies, and statistics.'

However,  In view of "An unsophisticated forecaster uses statistics like a drunken man uses a lamp post, more for support rather than illumination". one must be wary, as - Andrew Lang

"An approximate answer to the right problem is worth a good deal more than an exact answer to an approximate problem." -- John Tukey

All the researchers has contributed to their level best, and has got an idea that how i can validate the conceptual frame work. any how statisticians are statisticians they provide authentic proof through formula, later they support through a statement. i am a student of quantity and believe in these statistical tests. no one is wrong if he/she has done something through a valid procedure.

Muhammed,

I understand your question to be asking advice on writing up conclusions on your research. This usually involves connecting the results of the quantitative data analyses with the literature and providing implications for the field and recommendations for further research. I suggest reviewing your hypotheses with an eye toward the theoretical framework used to drive the question(s) of the study. Subsequent validation of construct validity for surveys requires showing discriminative validity such as between factors and between group and within groups in the sample data  along with data results the theoretical position in the nomological net. 

Construct validity is really developed through all aspects of the research though can often be supported through data correlation between similar constructs using other validated instruments run along with your own instrument data collection. I believe you understand these concepts as you have proven your competence through a successful defense of your study. 

Congratulations!

Janet

PS A next step is to run a study connecting your results (used as the independent variables) with a particular dependent variable in the educational leadership field. For example, often dependent variables in schools are academic achievement indicators such as accountability measures at the school level. In your case if ;you are studying Organizational Citizenship Behaviors of Principals an outcome measure could be improved school culture (using your choice of a variety of instruments). I used What's My School Mindset mean scores as an outcome measure for one of my studies.  Hope this helps,

Janet

Thanks a lot Janet,

 what you have suggested absolutely right, in first phase i did all the tests as you prescribed.  Face, content , discriminant  , and construct validity (AVE, CR values). Then i checked model fitness( absolute, incremental and parsimonious fit) CFI, RMSEA and Chi-Sq/df values. Now after all these i am also thinking of another study as suggested by you that how i can check the impact on students achievement in these secondary schools.

Your question is not that clear!

Are you talking about cross validation that is done by repeating the same experiment in a different place /time? Are you talking about internal -external validity of the design? Are you talking about construct validity? The first step in all of these is to go back to theory and build a theoretical model and after checking if your theoretical model is consistent with reality you have to check if it is consistent with previous research

 Dear Abdulrazzak,

your stance is right to some extent. Validity and reliability i have checked through respective indices, and i did it after valuable suggestions from various researchers like you. Your second point regarding theory, i designed my conceptual frame work on the basis of theoretical frame work, which is now found valid , because results support it. Now i think the problem is almost solved , now the next step is to check that whether this proposed frame work i s applicable in different context, which require another longitudinal study for its cross validation. Looking for further suggestions in this regard

I suggest going for cross validation :   repeating the same experiment in a different place /time and compare

Hi All

Thank you Janet Hanson

In addition to all answers,

A.Still, this unified concept of validity is best understood and examined within the context of its four discrete facets: content validity, construct validity, criterion validity and consequential validity.

B.Several authors suggest employing the following four steps to effectively evaluate content validity: 1) identify and outline the domain of interest, 2) gather resident domain experts, 3) develop consistent matching methodology, and 4) analyze results from the matching task.

C.Several series of steps in which to follow when pursuing a construct validation study: 1) generate hypotheses of how the construct should relate to both other constructs of interest and relevant group differences, 2) choose a measure that adequately represents the construct of interest, 3) pursue empirical study to examine the relationships hypothesized, and 4) analyze gathered data to check hypothesized relationships and to assess whether or not alternative hypotheses could explain the relationships found between the variables.

D.Criterion validity refers to the ability to draw accurate inferences from test scores to a related behavioral criterion of interest. This validity measure can be pursued in one of two contexts: predictive validity or concurrent validity. In predictive validity, researchers are interested in assessing the predictive utility of an instrument.

Please refer to,

Instrument Reliability and Validity:

Introductory Concepts and Measures

Amanda Jane Fairchild

James Madison University

https://libguides.lamar.edu/c.php?g=369302&p=3015608

Janet Hanson, Thanks for Answer.

following

Thanks a lot Abdelkader Mohamed Elsayed,

We can validate the conceptual framework for study in school leadership through the accuracy and relevance of the topics included within it.