My simple answer would be: absolutely not.

Multiplying your observations (or applying a weight factor of 7 to each observation) does not add any information to your dataset. Your analyses would assume that all observations are independent from each other, but that assumption would clearly be identified. 

Unfortunately, 30 observations is very limited from a statistical point of view. Probably the best way of proceeding is to use simple, bivariate statistics (correlations, cross-tables, etc.). With sample sizes that small, it would probably make sense to apply 'bootstrapping' (easy to do in recent versions of SPSS) to arrive at more valid confidence intervals (not being based on the central-limit theorem). 

My simple answer would be: absolutely not.

Multiplying your observations (or applying a weight factor of 7 to each observation) does not add any information to your dataset. Your analyses would assume that all observations are independent from each other, but that assumption would clearly be identified. 

Unfortunately, 30 observations is very limited from a statistical point of view. Probably the best way of proceeding is to use simple, bivariate statistics (correlations, cross-tables, etc.). With sample sizes that small, it would probably make sense to apply 'bootstrapping' (easy to do in recent versions of SPSS) to arrive at more valid confidence intervals (not being based on the central-limit theorem). 

I agree with "absolutely not". Also, although the bootstrap might free you from the assumption of a normal distribution, it will not reduce the sampling error inherent in the original sample.

In other words, if the original sample has a mean (or regression coefficient etc.) greater than that for the population, the bootstrap samples are also more likely to have the same.

Agree with the comments given above, and would like to add that if you generate more samples using bootstrap sampling, but the mean of bootstrap will still be same as original sample. 

I agree with the comments above. I will not multiply.

I will suggest you use the Partial Least Square Path Modelling (PLS-PM) approach. This is a type of SEM that can use few cases for modelling. See these articles:

Michel Tenenhaus et al (2005) .PLS Methodology to study the relationships between hedonic judgements and product characteristics, Food Quality and Preference, 16, 315-325.

Michel Tenehaus et al, (2005). PLS Path Modelling. Computational Statistics and Data Analysis, 48, 159-205.

I was not comfortable with the weighted premise. Thank you for the alternatives! 

I would say for 30 individuals, any type of inferential statistics is barely legit (you can run the analysis though). If I may extend, even descriptive statistics could have little to no value. In clinical research, a sample of 30 is still acceptable, but survey research with 30 individuals from a publishing perspective, has little value.

Understood Jagdish.  I have not had such little turnout on prior surveys but the conditions surrounding this one were hard to overcome in a foreign country and trouble at the borders.  But I think that it was a risk worth taking.  I am looking at alternative methods to increase participation including using some of my LinkedIn contacts to scrounge up legitimate participants.   But, I will have to include this in the methodology and am not sure how that will impact the impression of the results. May have to chalk this one as a learning experience. 

The validity of a sample size of 30 depends on the population being sampled. A poll for who you would vote for president of the United States with a sample size of 30 would be way too small. If you asked the same question but were only interested in hotel owners in Idaho, then 30 might be a perfectly valid sample because the population being examined is much smaller. Your population is already limited to those willing to take the survey and able to understand the questions being asked. How the survey was conducted may further limit the population.

If you read a journal article using the data that you have, would you trust the conclusions that you are reaching? Would you even be interested? If you were writing a review article (or a term paper) would this provide useful information that improves the understanding of other research or adds new context? If your results contradict all previous research, then the small sample size will be a big problem.

The issue reviewers will have to wrestle with is to answer a simple question: is this worth publishing given that it has insufficient sampling but also given that this is the best answer we are likely to get given current conditions?

I agree with previous comments: do not multiply your observations! Each survey is (in theory) an independent sample from the population of all possible individuals you are interested in. Copying the data does not give you more independent samples.

Tim: because I have done surveys with better than minimum sample requirements met, I am confident that I can use the analysis of 30 with some degree of precision since they will provide comparative value. Therefore, I believe I can make a case for it even though there are only 30 items.  As I have mentioned, I will be trying to utilize other resources and perhaps even redo the survey using the alternate method contacting myself instead of using 2 people in the country who may be strained because up upheaval in the area.  Though I agree it is important to weigh results on credibility, I am also aware of small sample size studies that can give initial impressions so that further studies can provide more detailed information.  Because this nation struggles with the same 'unknown impact' image for biomedical engineers as the US,, there may be value in simple analysis as a baseline since no studies in this area were performed in the US prior to mine and certainly none have been done in this Middle East nation, either.  All information is valuable to some extent, finding the right venue to publish may not be as hard as you say because all info in this area is basically novel and could therefore create interest.  At least, that is my hope.  Of course, the opposite can be true--some may not understand the overall hospital value and write it off but I certainly hope not. Thank you all for your comments.